20220824.md

ArXiv cs.CV --Wed, 24 Aug 2022

1.Distance-Aware Occlusion Detection with Focused Attention ⬇️

For humans, understanding the relationships between objects using visual signals is intuitive. For artificial intelligence, however, this task remains challenging. Researchers have made significant progress studying semantic relationship detection, such as human-object interaction detection and visual relationship detection. We take the study of visual relationships a step further from semantic to geometric. In specific, we predict relative occlusion and relative distance relationships. However, detecting these relationships from a single image is challenging. Enforcing focused attention to task-specific regions plays a critical role in successfully detecting these relationships. In this work, (1) we propose a novel three-decoder architecture as the infrastructure for focused attention; 2) we use the generalized intersection box prediction task to effectively guide our model to focus on occlusion-specific regions; 3) our model achieves a new state-of-the-art performance on distance-aware relationship detection. Specifically, our model increases the distance F1-score from 33.8% to 38.6% and boosts the occlusion F1-score from 34.4% to 41.2%. Our code is publicly available.

2.Towards Open Set Video Anomaly Detection ⬇️

Open Set Video Anomaly Detection (OpenVAD) aims to identify abnormal events from video data where both known anomalies and novel ones exist in testing. Unsupervised models learned solely from normal videos are applicable to any testing anomalies but suffer from a high false positive rate. In contrast, weakly supervised methods are effective in detecting known anomalies but could fail in an open world. We develop a novel weakly supervised method for the OpenVAD problem by integrating evidential deep learning (EDL) and normalizing flows (NFs) into a multiple instance learning (MIL) framework. Specifically, we propose to use graph neural networks and triplet loss to learn discriminative features for training the EDL classifier, where the EDL is capable of identifying the unknown anomalies by quantifying the uncertainty. Moreover, we develop an uncertainty-aware selection strategy to obtain clean anomaly instances and a NFs module to generate the pseudo anomalies. Our method is superior to existing approaches by inheriting the advantages of both the unsupervised NFs and the weakly-supervised MIL framework. Experimental results on multiple real-world video datasets show the effectiveness of our method.

3.DeepInteraction: 3D Object Detection via Modality Interaction ⬇️

Existing top-performance 3D object detectors typically rely on the multi-modal fusion strategy. This design is however fundamentally restricted due to overlooking the modality-specific useful information and finally hampering the model performance. To address this limitation, in this work we introduce a novel modality interaction strategy where individual per-modality representations are learned and maintained throughout for enabling their unique characteristics to be exploited during object detection. To realize this proposed strategy, we design a DeepInteraction architecture characterized by a multi-modal representational interaction encoder and a multi-modal predictive interaction decoder. Experiments on the large-scale nuScenes dataset show that our proposed method surpasses all prior arts often by a large margin. Crucially, our method is ranked at the first position at the highly competitive nuScenes object detection leaderboard.

4.Efficient Attention-free Video Shift Transformers ⬇️

This paper tackles the problem of efficient video recognition. In this area, video transformers have recently dominated the efficiency (top-1 accuracy vs FLOPs) spectrum. At the same time, there have been some attempts in the image domain which challenge the necessity of the self-attention operation within the transformer architecture, advocating the use of simpler approaches for token mixing. However, there are no results yet for the case of video recognition, where the self-attention operator has a significantly higher impact (compared to the case of images) on efficiency. To address this gap, in this paper, we make the following contributions: (a) we construct a highly efficient & accurate attention-free block based on the shift operator, coined Affine-Shift block, specifically designed to approximate as closely as possible the operations in the MHSA block of a Transformer layer. Based on our Affine-Shift block, we construct our Affine-Shift Transformer and show that it already outperforms all existing shift/MLP--based architectures for ImageNet classification. (b) We extend our formulation in the video domain to construct Video Affine-Shift Transformer (VAST), the very first purely attention-free shift-based video transformer. (c) We show that VAST significantly outperforms recent state-of-the-art transformers on the most popular action recognition benchmarks for the case of models with low computational and memory footprint. Code will be made available.

5.Explaining Bias in Deep Face Recognition via Image Characteristics ⬇️

In this paper, we propose a novel explanatory framework aimed to provide a better understanding of how face recognition models perform as the underlying data characteristics (protected attributes: gender, ethnicity, age; non-protected attributes: facial hair, makeup, accessories, face orientation and occlusion, image distortion, emotions) on which they are tested change. With our framework, we evaluate ten state-of-the-art face recognition models, comparing their fairness in terms of security and usability on two data sets, involving six groups based on gender and ethnicity. We then analyze the impact of image characteristics on models performance. Our results show that trends appearing in a single-attribute analysis disappear or reverse when multi-attribute groups are considered, and that performance disparities are also related to non-protected attributes. Source code: this https URL.

6.Consistency Regularization for Domain Adaptation ⬇️

Collection of real world annotations for training semantic segmentation models is an expensive process. Unsupervised domain adaptation (UDA) tries to solve this problem by studying how more accessible data such as synthetic data can be used to train and adapt models to real world images without requiring their annotations. Recent UDA methods applies self-learning by training on pixel-wise classification loss using a student and teacher network. In this paper, we propose the addition of a consistency regularization term to semi-supervised UDA by modelling the inter-pixel relationship between elements in networks' output. We demonstrate the effectiveness of the proposed consistency regularization term by applying it to the state-of-the-art DAFormer framework and improving mIoU19 performance on the GTA5 to Cityscapes benchmark by 0.8 and mIou16 performance on the SYNTHIA to Cityscapes benchmark by 1.2.

7.DepthFake: a depth-based strategy for detecting Deepfake videos ⬇️

Fake content has grown at an incredible rate over the past few years. The spread of social media and online platforms makes their dissemination on a large scale increasingly accessible by malicious actors. In parallel, due to the growing diffusion of fake image generation methods, many Deep Learning-based detection techniques have been proposed. Most of those methods rely on extracting salient features from RGB images to detect through a binary classifier if the image is fake or real. In this paper, we proposed DepthFake, a study on how to improve classical RGB-based approaches with depth-maps. The depth information is extracted from RGB images with recent monocular depth estimation techniques. Here, we demonstrate the effective contribution of depth-maps to the deepfake detection task on robust pre-trained architectures. The proposed RGBD approach is in fact able to achieve an average improvement of 3.20% and up to 11.7% for some deepfake attacks with respect to standard RGB architectures over the FaceForensic++ dataset.

8.Neuroevolution-based Classifiers for Deforestation Detection in Tropical Forests ⬇️

Tropical forests represent the home of many species on the planet for flora and fauna, retaining billions of tons of carbon footprint, promoting clouds and rain formation, implying a crucial role in the global ecosystem, besides representing the home to countless indigenous peoples. Unfortunately, millions of hectares of tropical forests are lost every year due to deforestation or degradation. To mitigate that fact, monitoring and deforestation detection programs are in use, in addition to public policies for the prevention and punishment of criminals. These monitoring/detection programs generally use remote sensing images, image processing techniques, machine learning methods, and expert photointerpretation to analyze, identify and quantify possible changes in forest cover. Several projects have proposed different computational approaches, tools, and models to efficiently identify recent deforestation areas, improving deforestation monitoring programs in tropical forests. In this sense, this paper proposes the use of pattern classifiers based on neuroevolution technique (NEAT) in tropical forest deforestation detection tasks. Furthermore, a novel framework called e-NEAT has been created and achieved classification results above $90%$ for balanced accuracy measure in the target application using an extremely reduced and limited training set for learning the classification models. These results represent a relative gain of $6.2%$ over the best baseline ensemble method compared in this paper

9.IMPaSh: A Novel Domain-shift Resistant Representation for Colorectal Cancer Tissue Classification ⬇️

The appearance of histopathology images depends on tissue type, staining and digitization procedure. These vary from source to source and are the potential causes for domain-shift problems. Owing to this problem, despite the great success of deep learning models in computational pathology, a model trained on a specific domain may still perform sub-optimally when we apply them to another domain. To overcome this, we propose a new augmentation called PatchShuffling and a novel self-supervised contrastive learning framework named IMPaSh for pre-training deep learning models. Using these, we obtained a ResNet50 encoder that can extract image representation resistant to domain-shift. We compared our derived representation against those acquired based on other domain-generalization techniques by using them for the cross-domain classification of colorectal tissue images. We show that the proposed method outperforms other traditional histology domain-adaptation and state-of-the-art self-supervised learning methods. Code is available at: this https URL .

10.Self-Trained Proposal Networks for the Open World ⬇️

Deep learning-based object proposal methods have enabled significant advances in many computer vision pipelines. However, current state-of-the-art proposal networks use a closed-world assumption, meaning they are only trained to detect instances of the training classes while treating every other region as background. This style of solution fails to provide high recall on out-of-distribution objects, rendering it inadequate for use in realistic open-world applications where novel object categories of interest may be observed. To better detect all objects, we propose a classification-free Self-Trained Proposal Network (STPN) that leverages a novel self-training optimization strategy combined with dynamically weighted loss functions that account for challenges such as class imbalance and pseudo-label uncertainty. Not only is our model designed to excel in existing optimistic open-world benchmarks, but also in challenging operating environments where there is significant label bias. To showcase this, we devise two challenges to test the generalization of proposal models when the training data contains (1) less diversity within the labeled classes, and (2) fewer labeled instances. Our results show that STPN achieves state-of-the-art novel object generalization on all tasks.

11.Flat Multi-modal Interaction Transformer for Named Entity Recognition ⬇️

Multi-modal named entity recognition (MNER) aims at identifying entity spans and recognizing their categories in social media posts with the aid of images. However, in dominant MNER approaches, the interaction of different modalities is usually carried out through the alternation of self-attention and cross-attention or over-reliance on the gating machine, which results in imprecise and biased correspondence between fine-grained semantic units of text and image. To address this issue, we propose a Flat Multi-modal Interaction Transformer (FMIT) for MNER. Specifically, we first utilize noun phrases in sentences and general domain words to obtain visual cues. Then, we transform the fine-grained semantic representation of the vision and text into a unified lattice structure and design a novel relative position encoding to match different modalities in Transformer. Meanwhile, we propose to leverage entity boundary detection as an auxiliary task to alleviate visual bias. Experiments show that our methods achieve the new state-of-the-art performance on two benchmark datasets.

12.CitySim: A Drone-Based Vehicle Trajectory Dataset for Safety Oriented Research and Digital Twins ⬇️

The development of safety-oriented research ideas and applications requires fine-grained vehicle trajectory data that not only has high accuracy but also captures a substantial number of critical safety events. This paper introduces the CitySim Dataset, which was devised with a core objective of facilitating safety-based research and applications. CitySim has vehicle trajectories extracted from 1140-minutes of drone videos recorded at 12 different locations. It covers a variety of road geometries including freeway basic segments, weaving segments, expressway merge/diverge segments, signalized intersections, stop-controlled intersections, and intersections without sign/signal control. CitySim trajectories were generated through a five-step procedure which ensured the trajectory accuracy. Furthermore, the dataset provides vehicle rotated bounding box information which is demonstrated to improve safety evaluation. Compared to other video-based trajectory datasets, the CitySim Dataset has significantly more critical safety events with higher severity including cut-in, merge, and diverge events. In addition, CitySim facilitates research towards digital twin applications by providing relevant assets like the recording locations'3D base maps and signal timings. These features enable more comprehensive conditions for safety research and applications such as autonomous vehicle safety and location-based safety analysis. The dataset is available online at this https URL.

13.Adversarial Feature Augmentation for Cross-domain Few-shot Classification ⬇️

Existing methods based on meta-learning predict novel-class labels for (target domain) testing tasks via meta knowledge learned from (source domain) training tasks of base classes. However, most existing works may fail to generalize to novel classes due to the probably large domain discrepancy across domains. To address this issue, we propose a novel adversarial feature augmentation (AFA) method to bridge the domain gap in few-shot learning. The feature augmentation is designed to simulate distribution variations by maximizing the domain discrepancy. During adversarial training, the domain discriminator is learned by distinguishing the augmented features (unseen domain) from the original ones (seen domain), while the domain discrepancy is minimized to obtain the optimal feature encoder. The proposed method is a plug-and-play module that can be easily integrated into existing few-shot learning methods based on meta-learning. Extensive experiments on nine datasets demonstrate the superiority of our method for cross-domain few-shot classification compared with the state of the art. Code is available at this https URL

14.Low-Light Video Enhancement with Synthetic Event Guidance ⬇️

Low-light video enhancement (LLVE) is an important yet challenging task with many applications such as photographing and autonomous driving. Unlike single image low-light enhancement, most LLVE methods utilize temporal information from adjacent frames to restore the color and remove the noise of the target frame. However, these algorithms, based on the framework of multi-frame alignment and enhancement, may produce multi-frame fusion artifacts when encountering extreme low light or fast motion. In this paper, inspired by the low latency and high dynamic range of events, we use synthetic events from multiple frames to guide the enhancement and restoration of low-light videos. Our method contains three stages: 1) event synthesis and enhancement, 2) event and image fusion, and 3) low-light enhancement. In this framework, we design two novel modules (event-image fusion transform and event-guided dual branch) for the second and third stages, respectively. Extensive experiments show that our method outperforms existing low-light video or single image enhancement approaches on both synthetic and real LLVE datasets.

15.AniWho : A Quick and Accurate Way to Classify Anime Character Faces in Images ⬇️

This paper aims to dive more deeply into various models available, including; InceptionV3, InceptionResNetV2, MobileNetV2, and EfficientNetB7 using transfer learning, to classify Japanese animation-style character faces. This paper has shown that EfficientNet-B7 has the highest accuracy rate with 85.08% top-1 Accuracy, followed by MobileNetV2, having a slightly less accurate result but with the benefits of much lower inference time and fewer number of required parameters. This paper also uses a few-shot learning framework, specifically Prototypical Networks, which produces decent results that can be used as an alternative to traditional transfer learning methods.

16.Adaptation of MobileNetV2 for Face Detection on Ultra-Low Power Platform ⬇️

Designing Deep Neural Networks (DNNs) running on edge hardware remains a challenge. Standard designs have been adopted by the community to facilitate the deployment of Neural Network models. However, not much emphasis is put on adapting the network topology to fit hardware constraints. In this paper, we adapt one of the most widely used architectures for mobile hardware platforms, MobileNetV2, and study the impact of changing its topology and applying post-training quantization. We discuss the impact of the adaptations and the deployment of the model on an embedded hardware platform for face detection.

17.An Evolutionary Approach for Creating of Diverse Classifier Ensembles ⬇️

Classification is one of the most studied tasks in data mining and machine learning areas and many works in the literature have been presented to solve classification problems for multiple fields of knowledge such as medicine, biology, security, and remote sensing. Since there is no single classifier that achieves the best results for all kinds of applications, a good alternative is to adopt classifier fusion strategies. A key point in the success of classifier fusion approaches is the combination of diversity and accuracy among classifiers belonging to an ensemble. With a large amount of classification models available in the literature, one challenge is the choice of the most suitable classifiers to compose the final classification system, which generates the need of classifier selection strategies. We address this point by proposing a framework for classifier selection and fusion based on a four-step protocol called CIF-E (Classifiers, Initialization, Fitness function, and Evolutionary algorithm). We implement and evaluate 24 varied ensemble approaches following the proposed CIF-E protocol and we are able to find the most accurate approach. A comparative analysis has also been performed among the best approaches and many other baselines from the literature. The experiments show that the proposed evolutionary approach based on Univariate Marginal Distribution Algorithm (UMDA) can outperform the state-of-the-art literature approaches in many well-known UCI datasets.

18.Quality Matters: Embracing Quality Clues for Robust 3D Multi-Object Tracking ⬇️

3D Multi-Object Tracking (MOT) has achieved tremendous achievement thanks to the rapid development of 3D object detection and 2D MOT. Recent advanced works generally employ a series of object attributes, e.g., position, size, velocity, and appearance, to provide the clues for the association in 3D MOT. However, these cues may not be reliable due to some visual noise, such as occlusion and blur, leading to tracking performance bottleneck. To reveal the dilemma, we conduct extensive empirical analysis to expose the key bottleneck of each clue and how they correlate with each other. The analysis results motivate us to efficiently absorb the merits among all cues, and adaptively produce an optimal tacking manner. Specifically, we present Location and Velocity Quality Learning, which efficiently guides the network to estimate the quality of predicted object attributes. Based on these quality estimations, we propose a quality-aware object association (QOA) strategy to leverage the quality score as an important reference factor for achieving robust association. Despite its simplicity, extensive experiments indicate that the proposed strategy significantly boosts tracking performance by 2.2% AMOTA and our method outperforms all existing state-of-the-art works on nuScenes by a large margin. Moreover, QTrack achieves 48.0% and 51.1% AMOTA tracking performance on the nuScenes validation and test sets, which significantly reduces the performance gap between pure camera and LiDAR based trackers.

19.Robust DNN Watermarking via Fixed Embedding Weights with Optimized Distribution ⬇️

Watermarking has been proposed as a way to protect the Intellectual Property Rights (IPR) of Deep Neural Networks (DNNs) and track their use. Several methods have been proposed that embed the watermark into the trainable parameters of the network (white box watermarking) or into the input-output mappping implemented by the network in correspondence to specific inputs (black box watermarking). In both cases, achieving robustness against fine tuning, model compression and, even more, transfer learning, is one of the most difficult challenges researchers are trying to face with. In this paper, we propose a new white-box, multi-bit watermarking algorithm with strong robustness properties, including retraining for transfer learning. Robustness is achieved thanks to a new information coding strategy according to which the watermark message is spread across a number of fixed weights, whose position depends on a secret key. The weights hosting the watermark are set prior to training, and are left unchanged throughout the entire training procedure. The distribution of the weights carrying out the message is theoretically optimised to make sure that the watermarked weights are indistinguishable from the other weights, while at the same time keeping their amplitude as large as possible to improve robustness against retraining. We carried out several experiments demonstrating the capability of the proposed scheme to provide high payloads with practically no impact on the network accuracy, at the same time retaining excellent robustness against network modifications an re-use, including retraining for transfer learning.

20.Doc-GCN: Heterogeneous Graph Convolutional Networks for Document Layout Analysis ⬇️

Recognizing the layout of unstructured digital documents is crucial when parsing the documents into the structured, machine-readable format for downstream applications. Recent studies in Document Layout Analysis usually rely on computer vision models to understand documents while ignoring other information, such as context information or relation of document components, which are vital to capture. Our Doc-GCN presents an effective way to harmonize and integrate heterogeneous aspects for Document Layout Analysis. We first construct graphs to explicitly describe four main aspects, including syntactic, semantic, density, and appearance/visual information. Then, we apply graph convolutional networks for representing each aspect of information and use pooling to integrate them. Finally, we aggregate each aspect and feed them into 2-layer MLPs for document layout component classification. Our Doc-GCN achieves new state-of-the-art results in three widely used DLA datasets.

21.PU-MFA : Point Cloud Up-sampling via Multi-scale Features Attention ⬇️

Recently, research using point clouds has been increasing with the development of 3D scanner technology. According to this trend, the demand for high-quality point clouds is increasing, but there is still a problem with the high cost of obtaining high-quality point clouds. Therefore, with the recent remarkable development of deep learning, point cloud up-sampling research, which uses deep learning to generate high-quality point clouds from low-quality point clouds, is one of the fields attracting considerable attention. This paper proposes a new point cloud up-sampling method called Point cloud Up-sampling via Multi-scale Features Attention (PU-MFA). Inspired by previous studies that reported good performance using the multi-scale features or attention mechanisms, PU-MFA merges the two through a U-Net structure. In addition, PU-MFA adaptively uses multi-scale features to refine the global features effectively. The performance of PU-MFA was compared with other state-of-the-art methods through various experiments using the PU-GAN dataset, which is a synthetic point cloud dataset, and the KITTI dataset, which is the real-scanned point cloud dataset. In various experimental results, PU-MFA showed superior performance in quantitative and qualitative evaluation compared to other state-of-the-art methods, proving the effectiveness of the proposed method. The attention map of PU-MFA was also visualized to show the effect of multi-scale features.

22.Deep Structural Causal Shape Models ⬇️

Causal reasoning provides a language to ask important interventional and counterfactual questions beyond purely statistical association. In medical imaging, for example, we may want to study the causal effect of genetic, environmental, or lifestyle factors on the normal and pathological variation of anatomical phenotypes. However, while anatomical shape models of 3D surface meshes, extracted from automated image segmentation, can be reliably constructed, there is a lack of computational tooling to enable causal reasoning about morphological variations. To tackle this problem, we propose deep structural causal shape models (CSMs), which utilise high-quality mesh generation techniques, from geometric deep learning, within the expressive framework of deep structural causal models. CSMs enable subject-specific prognoses through counterfactual mesh generation ("How would this patient's brain structure change if they were ten years older?"), which is in contrast to most current works on purely population-level statistical shape modelling. We demonstrate the capabilities of CSMs at all levels of Pearl's causal hierarchy through a number of qualitative and quantitative experiments leveraging a large dataset of 3D brain structures.

23.FS-BAN: Born-Again Networks for Domain Generalization Few-Shot Classification ⬇️

Conventional Few-shot classification (FSC) aims to recognize samples from novel classes given limited labeled data. Recently, domain generalization FSC (DG-FSC) has been proposed with the goal to recognize novel class samples from unseen domains. DG-FSC poses considerable challenges to many models due to the domain shift between base classes (used in training) and novel classes (encountered in evaluation). In this work, we make two novel contributions to tackle DG-FSC. Our first contribution is to propose Born-Again Network (BAN) episodic training and comprehensively investigate its effectiveness for DG-FSC. As a specific form of knowledge distillation, BAN has been shown to achieve improved generalization in conventional supervised classification with a closed-set setup. This improved generalization motivates us to study BAN for DG-FSC, and we show that BAN is promising to address the domain shift encountered in DG-FSC. Building on the encouraging finding, our second (major) contribution is to propose few-shot BAN, FS-BAN, a novel BAN approach for DG-FSC. Our proposed FS-BAN includes novel multi-task learning objectives: Mutual Regularization, Mismatched Teacher and Meta-Control Temperature, each of these is specifically designed to overcome central and unique challenges in DG-FSC, namely overfitting and domain discrepancy. We analyze different design choices of these techniques. We conduct comprehensive quantitative and qualitative analysis and evaluation using six datasets and three baseline models. The results suggest that our proposed FS-BAN consistently improves the generalization performance of baseline models and achieves state-of-the-art accuracy for DG-FSC.

24.Vox-Surf: Voxel-based Implicit Surface Representation ⬇️

Virtual content creation and interaction play an important role in modern 3D applications such as AR and VR. Recovering detailed 3D models from real scenes can significantly expand the scope of its applications and has been studied for decades in the computer vision and computer graphics community. We propose Vox-Surf, a voxel-based implicit surface representation. Our Vox-Surf divides the space into finite bounded voxels. Each voxel stores geometry and appearance information in its corner vertices. Vox-Surf is suitable for almost any scenario thanks to sparsity inherited from voxel representation and can be easily trained from multiple view images. We leverage the progressive training procedure to extract important voxels gradually for further optimization so that only valid voxels are preserved, which greatly reduces the number of sampling points and increases rendering speed.The fine voxels can also be considered as the bounding volume for collision detection.The experiments show that Vox-Surf representation can learn delicate surface details and accurate color with less memory and faster rendering speed than other methods.We also show that Vox-Surf can be more practical in scene editing and AR applications.

25.StyleTalker: One-shot Style-based Audio-driven Talking Head Video Generation ⬇️

We propose StyleTalker, a novel audio-driven talking head generation model that can synthesize a video of a talking person from a single reference image with accurately audio-synced lip shapes, realistic head poses, and eye blinks. Specifically, by leveraging a pretrained image generator and an image encoder, we estimate the latent codes of the talking head video that faithfully reflects the given audio. This is made possible with several newly devised components: 1) A contrastive lip-sync discriminator for accurate lip synchronization, 2) A conditional sequential variational autoencoder that learns the latent motion space disentangled from the lip movements, such that we can independently manipulate the motions and lip movements while preserving the identity. 3) An auto-regressive prior augmented with normalizing flow to learn a complex audio-to-motion multi-modal latent space. Equipped with these components, StyleTalker can generate talking head videos not only in a motion-controllable way when another motion source video is given but also in a completely audio-driven manner by inferring realistic motions from the input audio. Through extensive experiments and user studies, we show that our model is able to synthesize talking head videos with impressive perceptual quality which are accurately lip-synced with the input audios, largely outperforming state-of-the-art baselines.

26.Deepfake: Definitions, Performance Metrics and Standards, Datasets and Benchmarks, and a Meta-Review ⬇️

Recent advancements in AI, especially deep learning, have contributed to a significant increase in the creation of new realistic-looking synthetic media (video, image, and audio) and manipulation of existing media, which has led to the creation of the new term ``deepfake''. Based on both the research literature and resources in English and in Chinese, this paper gives a comprehensive overview of deepfake, covering multiple important aspects of this emerging concept, including 1) different definitions, 2) commonly used performance metrics and standards, and 3) deepfake-related datasets, challenges, competitions and benchmarks. In addition, the paper also reports a meta-review of 12 selected deepfake-related survey papers published in 2020 and 2021, focusing not only on the mentioned aspects, but also on the analysis of key challenges and recommendations. We believe that this paper is the most comprehensive review of deepfake in terms of aspects covered, and the first one covering both the English and Chinese literature and sources.

27.Can you recommend content to creatives instead of final consumers? A RecSys based on user's preferred visual styles ⬇️

Providing meaningful recommendations in a content marketplace is challenging due to the fact that users are not the final content consumers. Instead, most users are creatives whose interests, linked to the projects they work on, change rapidly and abruptly. To address the challenging task of recommending images to content creators, we design a RecSys that learns visual styles preferences transversal to the semantics of the projects users work on. We analyze the challenges of the task compared to content-based recommendations driven by semantics, propose an evaluation setup, and explain its applications in a global image marketplace.
This technical report is an extension of the paper "Learning Users' Preferred Visual Styles in an Image Marketplace", presented at ACM RecSys '22.

28.A Comprehensive Study of Real-Time Object Detection Networks Across Multiple Domains: A Survey ⬇️

Deep neural network based object detectors are continuously evolving and are used in a multitude of applications, each having its own set of requirements. While safety-critical applications need high accuracy and reliability, low-latency tasks need resource and energy-efficient networks. Real-time detectors, which are a necessity in high-impact real-world applications, are continuously proposed, but they overemphasize the improvements in accuracy and speed while other capabilities such as versatility, robustness, resource and energy efficiency are omitted. A reference benchmark for existing networks does not exist, nor does a standard evaluation guideline for designing new networks, which results in ambiguous and inconsistent comparisons. We, thus, conduct a comprehensive study on multiple real-time detectors (anchor-, keypoint-, and transformer-based) on a wide range of datasets and report results on an extensive set of metrics. We also study the impact of variables such as image size, anchor dimensions, confidence thresholds, and architecture layers on the overall performance. We analyze the robustness of detection networks against distribution shifts, natural corruptions, and adversarial attacks. Also, we provide a calibration analysis to gauge the reliability of the predictions. Finally, to highlight the real-world impact, we conduct two unique case studies, on autonomous driving and healthcare applications. To further gauge the capability of networks in critical real-time applications, we report the performance after deploying the detection networks on edge devices. Our extensive empirical study can act as a guideline for the industrial community to make an informed choice on the existing networks. We also hope to inspire the research community towards a new direction in the design and evaluation of networks that focuses on a bigger and holistic overview for a far-reaching impact.

29.FocusFormer: Focusing on What We Need via Architecture Sampler ⬇️

Vision Transformers (ViTs) have underpinned the recent breakthroughs in computer vision. However, designing the architectures of ViTs is laborious and heavily relies on expert knowledge. To automate the design process and incorporate deployment flexibility, one-shot neural architecture search decouples the supernet training and architecture specialization for diverse deployment scenarios. To cope with an enormous number of sub-networks in the supernet, existing methods treat all architectures equally important and randomly sample some of them in each update step during training. During architecture search, these methods focus on finding architectures on the Pareto frontier of performance and resource consumption, which forms a gap between training and deployment. In this paper, we devise a simple yet effective method, called FocusFormer, to bridge such a gap. To this end, we propose to learn an architecture sampler to assign higher sampling probabilities to those architectures on the Pareto frontier under different resource constraints during supernet training, making them sufficiently optimized and hence improving their performance. During specialization, we can directly use the well-trained architecture sampler to obtain accurate architectures satisfying the given resource constraint, which significantly improves the search efficiency. Extensive experiments on CIFAR-100 and ImageNet show that our FocusFormer is able to improve the performance of the searched architectures while significantly reducing the search cost. For example, on ImageNet, our FocusFormer-Ti with 1.4G FLOPs outperforms AutoFormer-Ti by 0.5% in terms of the Top-1 accuracy.

30.In-Air Imaging Sonar Sensor Network with Real-Time Processing Using GPUs ⬇️

For autonomous navigation and robotic applications, sensing the environment correctly is crucial. Many sensing modalities for this purpose exist. In recent years, one such modality that is being used is in-air imaging sonar. It is ideal in complex environments with rough conditions such as dust or fog. However, like with most sensing modalities, to sense the full environment around the mobile platform, multiple such sensors are needed to capture the full 360-degree range. Currently the processing algorithms used to create this data are insufficient to do so for multiple sensors at a reasonably fast update rate. Furthermore, a flexible and robust framework is needed to easily implement multiple imaging sonar sensors into any setup and serve multiple application types for the data. In this paper we present a sensor network framework designed for this novel sensing modality. Furthermore, an implementation of the processing algorithm on a Graphics Processing Unit is proposed to potentially decrease the computing time to allow for real-time processing of one or more imaging sonar sensors at a sufficiently high update rate.

31.Multimodal Crop Type Classification Fusing Multi-Spectral Satellite Time Series with Farmers Crop Rotations and Local Crop Distribution ⬇️

Accurate, detailed, and timely crop type mapping is a very valuable information for the institutions in order to create more accurate policies according to the needs of the citizens. In the last decade, the amount of available data dramatically increased, whether it can come from Remote Sensing (using Copernicus Sentinel-2 data) or directly from the farmers (providing in-situ crop information throughout the years and information on crop rotation). Nevertheless, the majority of the studies are restricted to the use of one modality (Remote Sensing data or crop rotation) and never fuse the Earth Observation data with domain knowledge like crop rotations. Moreover, when they use Earth Observation data they are mainly restrained to one year of data, not taking into account the past years. In this context, we propose to tackle a land use and crop type classification task using three data types, by using a Hierarchical Deep Learning algorithm modeling the crop rotations like a language model, the satellite signals like a speech signal and using the crop distribution as additional context vector. We obtained very promising results compared to classical approaches with significant performances, increasing the Accuracy by 5.1 points in a 28-class setting (.948), and the micro-F1 by 9.6 points in a 10-class setting (.887) using only a set of crop of interests selected by an expert. We finally proposed a data-augmentation technique to allow the model to classify the crop before the end of the season, which works surprisingly well in a multimodal setting.

32.Multimodal Across Domains Gaze Target Detection ⬇️

This paper addresses the gaze target detection problem in single images captured from the third-person perspective. We present a multimodal deep architecture to infer where a person in a scene is looking. This spatial model is trained on the head images of the person-of- interest, scene and depth maps representing rich context information. Our model, unlike several prior art, do not require supervision of the gaze angles, do not rely on head orientation information and/or location of the eyes of person-of-interest. Extensive experiments demonstrate the stronger performance of our method on multiple benchmark datasets. We also investigated several variations of our method by altering joint-learning of multimodal data. Some variations outperform a few prior art as well. First time in this paper, we inspect domain adaption for gaze target detection, and we empower our multimodal network to effectively handle the domain gap across datasets. The code of the proposed method is available at this https URL.

33.Towards Accurate Facial Landmark Detection via Cascaded Transformers ⬇️

Accurate facial landmarks are essential prerequisites for many tasks related to human faces. In this paper, an accurate facial landmark detector is proposed based on cascaded transformers. We formulate facial landmark detection as a coordinate regression task such that the model can be trained end-to-end. With self-attention in transformers, our model can inherently exploit the structured relationships between landmarks, which would benefit landmark detection under challenging conditions such as large pose and occlusion. During cascaded refinement, our model is able to extract the most relevant image features around the target landmark for coordinate prediction, based on deformable attention mechanism, thus bringing more accurate alignment. In addition, we propose a novel decoder that refines image features and landmark positions simultaneously. With few parameter increasing, the detection performance improves further. Our model achieves new state-of-the-art performance on several standard facial landmark detection benchmarks, and shows good generalization ability in cross-dataset evaluation.

34.Semantic Driven Energy based Out-of-Distribution Detection ⬇️

Detecting Out-of-Distribution (OOD) samples in real world visual applications like classification or object detection has become a necessary precondition in today's deployment of Deep Learning systems. Many techniques have been proposed, of which Energy based OOD methods have proved to be promising and achieved impressive performance. We propose semantic driven energy based method, which is an end-to-end trainable system and easy to optimize. We distinguish in-distribution samples from out-distribution samples with an energy score coupled with a representation score. We achieve it by minimizing the energy for in-distribution samples and simultaneously learn respective class representations that are closer and maximizing energy for out-distribution samples and pushing their representation further out from known class representation. Moreover, we propose a novel loss function which we call Cluster Focal Loss(CFL) that proved to be simple yet very effective in learning better class wise cluster center representations. We find that, our novel approach enhances outlier detection and achieve state-of-the-art as an energy-based model on common benchmarks. On CIFAR-10 and CIFAR-100 trained WideResNet, our model significantly reduces the relative average False Positive Rate(at True Positive Rate of 95%) by 67.2% and 57.4% respectively, compared to the existing energy based approaches. Further, we extend our framework for object detection and achieve improved performance.

35.Object Detection in Aerial Images with Uncertainty-Aware Graph Network ⬇️

In this work, we propose a novel uncertainty-aware object detection framework with a structured-graph, where nodes and edges are denoted by objects and their spatial-semantic similarities, respectively. Specifically, we aim to consider relationships among objects for effectively contextualizing them. To achieve this, we first detect objects and then measure their semantic and spatial distances to construct an object graph, which is then represented by a graph neural network (GNN) for refining visual CNN features for objects. However, refining CNN features and detection results of every object are inefficient and may not be necessary, as that include correct predictions with low uncertainties. Therefore, we propose to handle uncertain objects by not only transferring the representation from certain objects (sources) to uncertain objects (targets) over the directed graph, but also improving CNN features only on objects regarded as uncertain with their representational outputs from the GNN. Furthermore, we calculate a training loss by giving larger weights on uncertain objects, to concentrate on improving uncertain object predictions while maintaining high performances on certain objects. We refer to our model as Uncertainty-Aware Graph network for object DETection (UAGDet). We then experimentally validate ours on the challenging large-scale aerial image dataset, namely DOTA, that consists of lots of objects with small to large sizes in an image, on which ours improves the performance of the existing object detection network.

36.Efficient Self-Supervision using Patch-based Contrastive Learning for Histopathology Image Segmentation ⬇️

Learning discriminative representations of unlabelled data is a challenging task. Contrastive self-supervised learning provides a framework to learn meaningful representations using learned notions of similarity measures from simple pretext tasks. In this work, we propose a simple and efficient framework for self-supervised image segmentation using contrastive learning on image patches, without using explicit pretext tasks or any further labeled fine-tuning. A fully convolutional neural network (FCNN) is trained in a self-supervised manner to discern features in the input images and obtain confidence maps which capture the network's belief about the objects belonging to the same class. Positive- and negative- patches are sampled based on the average entropy in the confidence maps for contrastive learning. Convergence is assumed when the information separation between the positive patches is small, and the positive-negative pairs is large. We evaluate this method for the task of segmenting nuclei from multiple histopathology datasets, and show comparable performance with relevant self-supervised and supervised methods. The proposed model only consists of a simple FCNN with 10.8k parameters and requires about 5 minutes to converge on the high resolution microscopy datasets, which is orders of magnitude smaller than the relevant self-supervised methods to attain similar performance.

37.Adversarial Vulnerability of Temporal Feature Networks for Object Detection ⬇️

Taking into account information across the temporal domain helps to improve environment perception in autonomous driving. However, it has not been studied so far whether temporally fused neural networks are vulnerable to deliberately generated perturbations, i.e. adversarial attacks, or whether temporal history is an inherent defense against them. In this work, we study whether temporal feature networks for object detection are vulnerable to universal adversarial attacks. We evaluate attacks of two types: imperceptible noise for the whole image and locally-bound adversarial patch. In both cases, perturbations are generated in a white-box manner using PGD. Our experiments confirm, that attacking even a portion of a temporal input suffices to fool the network. We visually assess generated perturbations to gain insights into the functioning of attacks. To enhance the robustness, we apply adversarial training using 5-PGD. Our experiments on KITTI and nuScenes datasets demonstrate, that a model robustified via K-PGD is able to withstand the studied attacks while keeping the mAP-based performance comparable to that of an unattacked model.

38.Learning an Efficient Multimodal Depth Completion Model ⬇️

With the wide application of sparse ToF sensors in mobile devices, RGB image-guided sparse depth completion has attracted extensive attention recently, but still faces some problems. First, the fusion of multimodal information requires more network modules to process different modalities. But the application scenarios of sparse ToF measurements usually demand lightweight structure and low computational cost. Second, fusing sparse and noisy depth data with dense pixel-wise RGB data may introduce artifacts. In this paper, a light but efficient depth completion network is proposed, which consists of a two-branch global and local depth prediction module and a funnel convolutional spatial propagation network. The two-branch structure extracts and fuses cross-modal features with lightweight backbones. The improved spatial propagation module can refine the completed depth map gradually. Furthermore, corrected gradient loss is presented for the depth completion problem. Experimental results demonstrate the proposed method can outperform some state-of-the-art methods with a lightweight architecture. The proposed method also wins the championship in the MIPI2022 RGB+TOF depth completion challenge.

39.PIFu for the Real World: A Self-supervised Framework to Reconstruct Dressed Human from Single-view Images ⬇️

It is very challenging to accurately reconstruct sophisticated human geometry caused by various poses and garments from a single image. Recently, works based on pixel-aligned implicit function (PIFu) have made a big step and achieved state-of-the-art fidelity on image-based 3D human digitization. However, the training of PIFu relies heavily on expensive and limited 3D ground truth data (i.e. synthetic data), thus hindering its generalization to more diverse real world images. In this work, we propose an end-to-end self-supervised network named SelfPIFu to utilize abundant and diverse in-the-wild images, resulting in largely improved reconstructions when tested on unconstrained in-the-wild images. At the core of SelfPIFu is the depth-guided volume-/surface-aware signed distance fields (SDF) learning, which enables self-supervised learning of a PIFu without access to GT mesh. The whole framework consists of a normal estimator, a depth estimator, and a SDF-based PIFu and better utilizes extra depth GT during training. Extensive experiments demonstrate the effectiveness of our self-supervised framework and the superiority of using depth as input. On synthetic data, our Intersection-Over-Union (IoU) achieves to 93.5%, 18% higher compared with PIFuHD. For in-the-wild images, we conduct user studies on the reconstructed results, the selection rate of our results is over 68% compared with other state-of-the-art methods.

40.A Low-Cost Lane-Following Algorithm for Cyber-Physical Robots ⬇️

Duckiebots are low-cost mobile robots that are widely used in the fields of research and education. Although there are existing self-driving algorithms for the Duckietown platform, they are either too complex or perform too poorly to navigate a multi-lane track. Moreover, it is essential to give memory and computational resources to a Duckiebot so it can perform additional tasks such as out-of-distribution input detection. In order to satisfy these constraints, we built a low-cost autonomous driving algorithm capable of driving on a two-lane track. The algorithm uses traditional computer vision techniques to identify the central lane on the track and obtain the relevant steering angle. The steering is then controlled by a PID controller that smoothens the movement of the Duckiebot. The performance of the algorithm was compared to that of the NeurIPS 2018 AI Driving Olympics (AIDO) finalists, and it outperformed all but one finalists. The two main contributions of our algorithm are its low computational requirements and very quick set-up, with ongoing efforts to make it more reliable.

41.Depth Map Decomposition for Monocular Depth Estimation ⬇️

We propose a novel algorithm for monocular depth estimation that decomposes a metric depth map into a normalized depth map and scale features. The proposed network is composed of a shared encoder and three decoders, called G-Net, N-Net, and M-Net, which estimate gradient maps, a normalized depth map, and a metric depth map, respectively. M-Net learns to estimate metric depths more accurately using relative depth features extracted by G-Net and N-Net. The proposed algorithm has the advantage that it can use datasets without metric depth labels to improve the performance of metric depth estimation. Experimental results on various datasets demonstrate that the proposed algorithm not only provides competitive performance to state-of-the-art algorithms but also yields acceptable results even when only a small amount of metric depth data is available for its training.

42.CRCNet: Few-shot Segmentation with Cross-Reference and Region-Global Conditional Networks ⬇️

Few-shot segmentation aims to learn a segmentation model that can be generalized to novel classes with only a few training images. In this paper, we propose a Cross-Reference and Local-Global Conditional Networks (CRCNet) for few-shot segmentation. Unlike previous works that only predict the query image's mask, our proposed model concurrently makes predictions for both the support image and the query image. Our network can better find the co-occurrent objects in the two images with a cross-reference mechanism, thus helping the few-shot segmentation task. To further improve feature comparison, we develop a local-global conditional module to capture both global and local relations. We also develop a mask refinement module to refine the prediction of the foreground regions recurrently. Experiments on the PASCAL VOC 2012, MS COCO, and FSS-1000 datasets show that our network achieves new state-of-the-art performance.

43.How good are deep models in understanding\ the generated images? ⬇️

My goal in this paper is twofold: to study how well deep models can understand the images generated by DALL-E 2 and Midjourney, and to quantitatively evaluate these generative models. Two sets of generated images are collected for object recognition and visual question answering (VQA) tasks. On object recognition, the best model, out of 10 state-of-the-art object recognition models, achieves about 60% and 80% top-1 and top-5 accuracy, respectively. These numbers are much lower than the best accuracy on the ImageNet dataset (91% and 99%). On VQA, the OFA model scores 77.3% on answering 241 binary questions across 50 images. This model scores 94.7% on the binary VQA-v2 dataset. Humans are able to recognize the generated images and answer questions on them easily. We conclude that a) deep models struggle to understand the generated content, and may do better after fine-tuning, and b) there is a large distribution shift between the generated images and the real photographs. The distribution shift appears to be category-dependent. Data is available at: this https URL.

44.Learning More May Not Be Better: Knowledge Transferability in Vision and Language Tasks ⬇️

Is more data always better to train vision-and-language models? We study knowledge transferability in multi-modal tasks. The current tendency in machine learning is to assume that by joining multiple datasets from different tasks their overall performance will improve. However, we show that not all the knowledge transfers well or has a positive impact on related tasks, even when they share a common goal. We conduct an exhaustive analysis based on hundreds of cross-experiments on 12 vision-and-language tasks categorized in 4 groups. Whereas tasks in the same group are prone to improve each other, results show that this is not always the case. Other factors such as dataset size or pre-training stage have also a great impact on how well the knowledge is transferred.

45.Neural PCA for Flow-Based Representation Learning ⬇️

Of particular interest is to discover useful representations solely from observations in an unsupervised generative manner. However, the question of whether existing normalizing flows provide effective representations for downstream tasks remains mostly unanswered despite their strong ability for sample generation and density estimation. This paper investigates this problem for such a family of generative models that admits exact invertibility. We propose Neural Principal Component Analysis (Neural-PCA) that operates in full dimensionality while capturing principal components in \emph{descending} order. Without exploiting any label information, the principal components recovered store the most informative elements in their \emph{leading} dimensions and leave the negligible in the \emph{trailing} ones, allowing for clear performance improvements of $5%$-$10%$ in downstream tasks. Such improvements are empirically found consistent irrespective of the number of latent trailing dimensions dropped. Our work suggests that necessary inductive bias be introduced into generative modelling when representation quality is of interest.

46.Digital topological groups ⬇️

In this article, we develop the basic theory of digital topological groups. The basic definitions directly lead to two separate categories, based on the details of the continuity required of the group multiplication. We define $\NP_1$- and $\NP_2$-digital topological groups, and investigate their properties and algebraic structure. The $\NP_2$ category is very restrictive, and we give a complete classification of $\NP_2$-digital topological groups. We also give many examples of $\NP_1$-digital topological groups. We define digital topological group homomorphisms, and describe the digital counterpart of the first isomorphism theorem.

47.Hierarchically Decomposed Graph Convolutional Networks for Skeleton-Based Action Recognition ⬇️

Graph convolutional networks (GCNs) are the most commonly used method for skeleton-based action recognition and have achieved remarkable performance. Generating adjacency matrices with semantically meaningful edges is particularly important for this task, but extracting such edges is challenging problem. To solve this, we propose a hierarchically decomposed graph convolutional network (HD-GCN) architecture with a novel hierarchically decomposed graph (HD-Graph). The proposed HD-GCN effectively decomposes every joint node into several sets to extract major adjacent and distant edges, and uses them to construct an HD-Graph containing those edges in the same semantic spaces of a human skeleton. In addition, we introduce an attention-guided hierarchy aggregation (A-HA) module to highlight the dominant hierarchical edge sets of the HD-Graph. Furthermore, we apply a new two-stream-three-graph ensemble method, which uses only joint and bone stream without any motion stream. The proposed model is evaluated and achieves state-of-the-art performance on three large, popular datasets: NTU-RGB+D 60, NTU-RGB+D 120, and Northwestern-UCLA. Finally, we demonstrate the effectiveness of our model with various comparative experiments.

48.Quality-Constant Per-Shot Encoding by Two-Pass Learning-based Rate Factor Prediction ⬇️

Providing quality-constant streams can simultaneously guarantee user experience and prevent wasting bit-rate. In this paper, we propose a novel deep learning based two-pass encoder parameter prediction framework to decide rate factor (RF), with which encoder can output streams with constant quality. For each one-shot segment in a video, the proposed method firstly extracts spatial, temporal and pre-coding features by an ultra fast pre-process. Based on these features, a RF parameter is predicted by a deep neural network. Video encoder uses the RF to compress segment as the first encoding pass. Then VMAF quality of the first pass encoding is measured. If the quality doesn't meet target, a second pass RF prediction and encoding will be performed. With the help of first pass predicted RF and corresponding actual quality as feedback, the second pass prediction will be highly accurate. Experiments show the proposed method requires only 1.55 times encoding complexity on average, meanwhile the accuracy, that the compressed video's actual VMAF is within $\pm1$ around the target VMAF, reaches 98.88%.

49.Super-resolution 3D Human Shape from a Single Low-Resolution Image ⬇️

We propose a novel framework to reconstruct super-resolution human shape from a single low-resolution input image. The approach overcomes limitations of existing approaches that reconstruct 3D human shape from a single image, which require high-resolution images together with auxiliary data such as surface normal or a parametric model to reconstruct high-detail shape. The proposed framework represents the reconstructed shape with a high-detail implicit function. Analogous to the objective of 2D image super-resolution, the approach learns the mapping from a low-resolution shape to its high-resolution counterpart and it is applied to reconstruct 3D shape detail from low-resolution images. The approach is trained end-to-end employing a novel loss function which estimates the information lost between a low and high-resolution representation of the same 3D surface shape. Evaluation for single image reconstruction of clothed people demonstrates that our method achieves high-detail surface reconstruction from low-resolution images without auxiliary data. Extensive experiments show that the proposed approach can estimate super-resolution human geometries with a significantly higher level of detail than that obtained with previous approaches when applied to low-resolution images.

50.Bag of Tricks for Out-of-Distribution Generalization ⬇️

Recently, out-of-distribution (OOD) generalization has attracted attention to the robustness and generalization ability of deep learning based models, and accordingly, many strategies have been made to address different aspects related to this issue. However, most existing algorithms for OOD generalization are complicated and specifically designed for certain dataset. To alleviate this problem, nicochallenge-2022 provides NICO++, a large-scale dataset with diverse context information. In this paper, based on systematic analysis of different schemes on NICO++ dataset, we propose a simple but effective learning framework via coupling bag of tricks, including multi-objective framework design, data augmentations, training and inference strategies. Our algorithm is memory-efficient and easily-equipped, without complicated modules and does not require for large pre-trained models. It achieves an excellent performance with Top-1 accuracy of 88.16% on public test set and 75.65% on private test set, and ranks 1st in domain generalization task of nicochallenge-2022.

51.Threshold-adaptive Unsupervised Focal Loss for Domain Adaptation of Semantic Segmentation ⬇️

Semantic segmentation is an important task for intelligent vehicles to understand the environment. Current deep learning methods require large amounts of labeled data for training. Manual annotation is expensive, while simulators can provide accurate annotations. However, the performance of the semantic segmentation model trained with the data of the simulator will significantly decrease when applied in the actual scene. Unsupervised domain adaptation (UDA) for semantic segmentation has recently gained increasing research attention, aiming to reduce the domain gap and improve the performance on the target domain. In this paper, we propose a novel two-stage entropy-based UDA method for semantic segmentation. In stage one, we design a threshold-adaptative unsupervised focal loss to regularize the prediction in the target domain, which has a mild gradient neutralization mechanism and mitigates the problem that hard samples are barely optimized in entropy-based methods. In stage two, we introduce a data augmentation method named cross-domain image mixing (CIM) to bridge the semantic knowledge from two domains. Our method achieves state-of-the-art 58.4% and 59.6% mIoUs on SYNTHIA-to-Cityscapes and GTA5-to-Cityscapes using DeepLabV2 and competitive performance using the lightweight BiSeNet.

52.A Constrained Deformable Convolutional Network for Efficient Single Image Dynamic Scene Blind Deblurring with Spatially-Variant Motion Blur Kernels Estimation ⬇️

Most existing deep-learning-based single image dynamic scene blind deblurring (SIDSBD) methods usually design deep networks to directly remove the spatially-variant motion blurs from one inputted motion blurred image, without blur kernels estimation. In this paper, inspired by the Projective Motion Path Blur (PMPB) model and deformable convolution, we propose a novel constrained deformable convolutional network (CDCN) for efficient single image dynamic scene blind deblurring, which simultaneously achieves accurate spatially-variant motion blur kernels estimation and the high-quality image restoration from only one observed motion blurred image. In our proposed CDCN, we first construct a novel multi-scale multi-level multi-input multi-output (MSML-MIMO) encoder-decoder architecture for more powerful features extraction ability. Second, different from the DLVBD methods that use multiple consecutive frames, a novel constrained deformable convolution reblurring (CDCR) strategy is proposed, in which the deformable convolution is first applied to blurred features of the inputted single motion blurred image for learning the sampling points of motion blur kernel of each pixel, which is similar to the estimation of the motion density function of the camera shake in the PMPB model, and then a novel PMPB-based reblurring loss function is proposed to constrain the learned sampling points convergence, which can make the learned sampling points match with the relative motion trajectory of each pixel better and promote the accuracy of the spatially-variant motion blur kernels estimation.

53.Faint Features Tell: Automatic Vertebrae Fracture Screening Assisted by Contrastive Learning ⬇️

Long-term vertebral fractures severely affect the life quality of patients, causing kyphotic, lumbar deformity and even paralysis. Computed tomography (CT) is a common clinical examination to screen for this disease at early stages. However, the faint radiological appearances and unspecific symptoms lead to a high risk of missed diagnosis. In particular, the mild fractures and normal controls are quite difficult to distinguish for deep learning models and inexperienced doctors. In this paper, we argue that reinforcing the faint fracture features to encourage the inter-class separability is the key to improving the accuracy. Motivated by this, we propose a supervised contrastive learning based model to estimate Genent's Grade of vertebral fracture with CT scans. The supervised contrastive learning, as an auxiliary task, narrows the distance of features within the same class while pushing others away, which enhances the model's capability of capturing subtle features of vertebral fractures. Considering the lack of datasets in this field, we construct a database including 208 samples annotated by experienced radiologists. Our method has a specificity of 99% and a sensitivity of 85% in binary classification, and a macio-F1 of 77% in multi-classification, indicating that contrastive learning significantly improves the accuracy of vertebrae fracture screening, especially for the mild fractures and normal controls. Our desensitized data and codes will be made publicly available for the community.

54.Structure Regularized Attentive Network for Automatic Femoral Head Necrosis Diagnosis and Localization ⬇️

In recent years, several works have adopted the convolutional neural network (CNN) to diagnose the avascular necrosis of the femoral head (AVNFH) based on X-ray images or magnetic resonance imaging (MRI). However, due to the tissue overlap, X-ray images are difficult to provide fine-grained features for early diagnosis. MRI, on the other hand, has a long imaging time, is more expensive, making it impractical in mass screening. Computed tomography (CT) shows layer-wise tissues, is faster to image, and is less costly than MRI. However, to our knowledge, there is no work on CT-based automated diagnosis of AVNFH. In this work, we collected and labeled a large-scale dataset for AVNFH ranking. In addition, existing end-to-end CNNs only yields the classification result and are difficult to provide more information for doctors in diagnosis. To address this issue, we propose the structure regularized attentive network (SRANet), which is able to highlight the necrotic regions during classification based on patch attention. SRANet extracts features in chunks of images, obtains weight via the attention mechanism to aggregate the features, and constrains them by a structural regularizer with prior knowledge to improve the generalization. SRANet was evaluated on our AVNFH-CT dataset. Experimental results show that SRANet is superior to CNNs for AVNFH classification, moreover, it can localize lesions and provide more information to assist doctors in diagnosis. Our codes are made public at this https URL.

55.Spiral Contrastive Learning: An Efficient 3D Representation Learning Method for Unannotated CT Lesions ⬇️

Computed tomography (CT) samples with pathological annotations are difficult to obtain. As a result, the computer-aided diagnosis (CAD) algorithms are trained on small datasets (e.g., LIDC-IDRI with 1,018 samples), limiting their accuracies and reliability. In the past five years, several works have tailored for unsupervised representations of CT lesions via two-dimensional (2D) and three-dimensional (3D) self-supervised learning (SSL) algorithms. The 2D algorithms have difficulty capturing 3D information, and existing 3D algorithms are computationally heavy. Light-weight 3D SSL remains the boundary to explore. In this paper, we propose the spiral contrastive learning (SCL), which yields 3D representations in a computationally efficient manner. SCL first transforms 3D lesions to the 2D plane using an information-preserving spiral transformation, and then learn transformation-invariant features using 2D contrastive learning. For the augmentation, we consider natural image augmentations and medical image augmentations. We evaluate SCL by training a classification head upon the embedding layer. Experimental results show that SCL achieves state-of-the-art accuracy on LIDC-IDRI (89.72%), LNDb (82.09%) and TianChi (90.16%) for unsupervised representation learning. With 10% annotated data for fine-tune, the performance of SCL is comparable to that of supervised learning algorithms (85.75% vs. 85.03% on LIDC-IDRI, 78.20% vs. 73.44% on LNDb and 87.85% vs. 83.34% on TianChi, respectively). Meanwhile, SCL reduces the computational effort by 66.98% compared to other 3D SSL algorithms, demonstrating the efficiency of the proposed method in unsupervised pre-training.

56.Hierarchical Perceptual Noise Injection for Social Media Fingerprint Privacy Protection ⬇️

Billions of people are sharing their daily life images on social media every day. However, their biometric information (e.g., fingerprint) could be easily stolen from these images. The threat of fingerprint leakage from social media raises a strong desire for anonymizing shared images while maintaining image qualities, since fingerprints act as a lifelong individual biometric password. To guard the fingerprint leakage, adversarial attack emerges as a solution by adding imperceptible perturbations on images. However, existing works are either weak in black-box transferability or appear unnatural. Motivated by visual perception hierarchy (i.e., high-level perception exploits model-shared semantics that transfer well across models while low-level perception extracts primitive stimulus and will cause high visual sensitivities given suspicious stimulus), we propose FingerSafe, a hierarchical perceptual protective noise injection framework to address the mentioned problems. For black-box transferability, we inject protective noises on fingerprint orientation field to perturb the model-shared high-level semantics (i.e., fingerprint ridges). Considering visual naturalness, we suppress the low-level local contrast stimulus by regularizing the response of Lateral Geniculate Nucleus. Our FingerSafe is the first to provide feasible fingerprint protection in both digital (up to 94.12%) and realistic scenarios (Twitter and Facebook, up to 68.75%). Our code can be found at this https URL.

57.Learning from Noisy Labels with Coarse-to-Fine Sample Credibility Modeling ⬇️

Training deep neural network (DNN) with noisy labels is practically challenging since inaccurate labels severely degrade the generalization ability of DNN. Previous efforts tend to handle part or full data in a unified denoising flow via identifying noisy data with a coarse small-loss criterion to mitigate the interference from noisy labels, ignoring the fact that the difficulties of noisy samples are different, thus a rigid and unified data selection pipeline cannot tackle this problem well. In this paper, we first propose a coarse-to-fine robust learning method called CREMA, to handle noisy data in a divide-and-conquer manner. In coarse-level, clean and noisy sets are firstly separated in terms of credibility in a statistical sense. Since it is practically impossible to categorize all noisy samples correctly, we further process them in a fine-grained manner via modeling the credibility of each sample. Specifically, for the clean set, we deliberately design a memory-based modulation scheme to dynamically adjust the contribution of each sample in terms of its historical credibility sequence during training, thus alleviating the effect from noisy samples incorrectly grouped into the clean set. Meanwhile, for samples categorized into the noisy set, a selective label update strategy is proposed to correct noisy labels while mitigating the problem of correction error. Extensive experiments are conducted on benchmarks of different modalities, including image classification (CIFAR, Clothing1M etc) and text recognition (IMDB), with either synthetic or natural semantic noises, demonstrating the superiority and generality of CREMA.

58.Unsupervised Fish Trajectory Tracking and Segmentation ⬇️

DNN for fish tracking and segmentation based on high-quality labels is expensive. Alternative unsupervised approaches rely on spatial and temporal variations that naturally occur in video data to generate noisy pseudo-ground-truth labels. These pseudo-labels are used to train a multi-task deep neural network. In this paper, we propose a three-stage framework for robust fish tracking and segmentation, where the first stage is an optical flow model, which generates the pseudo labels using spatial and temporal consistency between frames. In the second stage, a self-supervised model refines the pseudo-labels incrementally. In the third stage, the refined labels are used to train a segmentation network. No human annotations are used during the training or inference. Extensive experiments are performed to validate our method on three public underwater video datasets and to demonstrate that it is highly effective for video annotation and segmentation. We also evaluate the robustness of our framework to different imaging conditions and discuss the limitations of our current implementation.

59.A First Look at Dataset Bias in License Plate Recognition ⬇️

Public datasets have played a key role in advancing the state of the art in License Plate Recognition (LPR). Although dataset bias has been recognized as a severe problem in the computer vision community, it has been largely overlooked in the LPR literature. LPR models are usually trained and evaluated separately on each dataset. In this scenario, they have often proven robust in the dataset they were trained in but showed limited performance in unseen ones. Therefore, this work investigates the dataset bias problem in the LPR context. We performed experiments on eight datasets, four collected in Brazil and four in mainland China, and observed that each dataset has a unique, identifiable "signature" since a lightweight classification model predicts the source dataset of a license plate (LP) image with more than 95% accuracy. In our discussion, we draw attention to the fact that most LPR models are probably exploiting such signatures to improve the results achieved in each dataset at the cost of losing generalization capability. These results emphasize the importance of evaluating LPR models in cross-dataset setups, as they provide a better indication of generalization (hence real-world performance) than within-dataset ones.

60.Learning Visibility for Robust Dense Human Body Estimation ⬇️

Estimating 3D human pose and shape from 2D images is a crucial yet challenging task. While prior methods with model-based representations can perform reasonably well on whole-body images, they often fail when parts of the body are occluded or outside the frame. Moreover, these results usually do not faithfully capture the human silhouettes due to their limited representation power of deformable models (e.g., representing only the naked body). An alternative approach is to estimate dense vertices of a predefined template body in the image space. Such representations are effective in localizing vertices within an image but cannot handle out-of-frame body parts. In this work, we learn dense human body estimation that is robust to partial observations. We explicitly model the visibility of human joints and vertices in the x, y, and z axes separately. The visibility in x and y axes help distinguishing out-of-frame cases, and the visibility in depth axis corresponds to occlusions (either self-occlusions or occlusions by other objects). We obtain pseudo ground-truths of visibility labels from dense UV correspondences and train a neural network to predict visibility along with 3D coordinates. We show that visibility can serve as 1) an additional signal to resolve depth ordering ambiguities of self-occluded vertices and 2) a regularization term when fitting a human body model to the predictions. Extensive experiments on multiple 3D human datasets demonstrate that visibility modeling significantly improves the accuracy of human body estimation, especially for partial-body cases. Our project page with code is at: this https URL.

61.Anatomy-Aware Contrastive Representation Learning for Fetal Ultrasound ⬇️

Self-supervised contrastive representation learning offers the advantage of learning meaningful visual representations from unlabeled medical datasets for transfer learning. However, applying current contrastive learning approaches to medical data without considering its domain-specific anatomical characteristics may lead to visual representations that are inconsistent in appearance and semantics. In this paper, we propose to improve visual representations of medical images via anatomy-aware contrastive learning (AWCL), which incorporates anatomy information to augment the positive/negative pair sampling in a contrastive learning manner. The proposed approach is demonstrated for automated fetal ultrasound imaging tasks, enabling the positive pairs from the same or different ultrasound scans that are anatomically similar to be pulled together and thus improving the representation learning. We empirically investigate the effect of inclusion of anatomy information with coarse- and fine-grained granularity, for contrastive learning and find that learning with fine-grained anatomy information which preserves intra-class difference is more effective than its counterpart. We also analyze the impact of anatomy ratio on our AWCL framework and find that using more distinct but anatomically similar samples to compose positive pairs results in better quality representations. Experiments on a large-scale fetal ultrasound dataset demonstrate that our approach is effective for learning representations that transfer well to three clinical downstream tasks, and achieves superior performance compared to ImageNet supervised and the current state-of-the-art contrastive learning methods. In particular, AWCL outperforms ImageNet supervised method by 13.8% and state-of-the-art contrastive-based method by 7.1% on a cross-domain segmentation task.

62.Individual Tree Detection in Large-Scale Urban Environments using High-Resolution Multispectral Imagery ⬇️

We introduce a novel deep learning method for detection of individual trees in urban environments using high-resolution multispectral aerial imagery. We use a convolutional neural network to regress a confidence map indicating the locations of individual trees, which are localized using a peak finding algorithm. Our method provides complete spatial coverage by detecting trees in both public and private spaces, and can scale to very large areas. In our study area spanning five cities in Southern California, we achieved an F-score of 0.735 and an RMSE of 2.157 m. We used our method to produce a map of all trees in the urban forest of California, indicating the potential for our method to support future urban forestry studies at unprecedented scales.

63.Automated Temporal Segmentation of Orofacial Assessment Videos ⬇️

Computer vision techniques can help automate or partially automate clinical examination of orofacial impairments to provide accurate and objective assessments. Towards the development of such automated systems, we evaluated two approaches to detect and temporally segment (parse) repetitions in orofacial assessment videos. Recorded videos of participants with amyotrophic lateral sclerosis (ALS) and healthy control (HC) individuals were obtained from the Toronto NeuroFace Dataset. Two approaches for repetition detection and parsing were examined: one based on engineered features from tracked facial landmarks and peak detection in the distance between the vermilion-cutaneous junction of the upper and lower lips (baseline analysis), and another using a pre-trained transformer-based deep learning model called RepNet (Dwibedi et al, 2020), which automatically detects periodicity, and parses periodic and semi-periodic repetitions in video data. In experimental evaluation of two orofacial assessments tasks, - repeating maximum mouth opening (OPEN) and repeating the sentence "Buy Bobby a Puppy" (BBP) - RepNet provided better parsing than the landmark-based approach, quantified by higher mean intersection-over-union (IoU) with respect to ground truth manual parsing. Automated parsing using RepNet also clearly separated HC and ALS participants based on the duration of BBP repetitions, whereas the landmark-based method could not.

64.A Survey and Framework of Cooperative Perception: From Heterogeneous Singleton to Hierarchical Cooperation ⬇️

Perceiving the environment is one of the most fundamental keys to enabling Cooperative Driving Automation (CDA), which is regarded as the revolutionary solution to addressing the safety, mobility, and sustainability issues of contemporary transportation systems. Although an unprecedented evolution is now happening in the area of computer vision for object perception, state-of-the-art perception methods are still struggling with sophisticated real-world traffic environments due to the inevitably physical occlusion and limited receptive field of single-vehicle systems. Based on multiple spatially separated perception nodes, Cooperative Perception (CP) is born to unlock the bottleneck of perception for driving automation. In this paper, we comprehensively review and analyze the research progress on CP and, to the best of our knowledge, this is the first time to propose a unified CP framework. Architectures and taxonomy of CP systems based on different types of sensors are reviewed to show a high-level description of the workflow and different structures for CP systems. Node structure, sensor modality, and fusion schemes are reviewed and analyzed with comprehensive literature to provide detailed explanations of specific methods. A Hierarchical CP framework is proposed, followed by a review of existing Datasets and Simulators to sketch an overall landscape of CP. Discussion highlights the current opportunities, open challenges, and anticipated future trends.

65.EBSnoR: Event-Based Snow Removal by Optimal Dwell Time Thresholding ⬇️

We propose an Event-Based Snow Removal algorithm called EBSnoR. We developed a technique to measure the dwell time of snowflakes on a pixel using event-based camera data, which is used to carry out a Neyman-Pearson hypothesis test to partition event stream into snowflake and background events. The effectiveness of the proposed EBSnoR was verified on a new dataset called UDayton22EBSnow, comprised of front-facing event-based camera in a car driving through snow with manually annotated bounding boxes around surrounding vehicles. Qualitatively, EBSnoR correctly identifies events corresponding to snowflakes; and quantitatively, EBSnoR-preprocessed event data improved the performance of event-based car detection algorithms.

66.State Of The Art In Open-Set Iris Presentation Attack Detection ⬇️

Research in presentation attack detection (PAD) for iris recognition has largely moved beyond evaluation in "closed-set" scenarios, to emphasize ability to generalize to presentation attack types not present in the training data. This paper offers several contributions to understand and extend the state-of-the-art in open-set iris PAD. First, it describes the most authoritative evaluation to date of iris PAD. We have curated the largest publicly-available image dataset for this problem, drawing from 26 benchmarks previously released by various groups, and adding 150,000 images being released with the journal version of this paper, to create a set of 450,000 images representing authentic iris and seven types of presentation attack instrument (PAI). We formulate a leave-one-PAI-out evaluation protocol, and show that even the best algorithms in the closed-set evaluations exhibit catastrophic failures on multiple attack types in the open-set scenario. This includes algorithms performing well in the most recent LivDet-Iris 2020 competition, which may come from the fact that the LivDet-Iris protocol emphasizes sequestered images rather than unseen attack types. Second, we evaluate the accuracy of five open-source iris presentation attack algorithms available today, one of which is newly-proposed in this paper, and build an ensemble method that beats the winner of the LivDet-Iris 2020 by a substantial margin. This paper demonstrates that closed-set iris PAD, when all PAIs are known during training, is a solved problem, with multiple algorithms showing very high accuracy, while open-set iris PAD, when evaluated correctly, is far from being solved. The newly-created dataset, new open-source algorithms, and evaluation protocol, made publicly available with the journal version of this paper, provide the experimental artifacts that researchers can use to measure progress on this important problem.

67.Transductive Decoupled Variational Inference for Few-Shot Classification ⬇️

The versatility to learn from a handful of samples is the hallmark of human intelligence. Few-shot learning is an endeavour to transcend this capability down to machines. Inspired by the promise and power of probabilistic deep learning, we propose a novel variational inference network for few-shot classification (coined as TRIDENT) to decouple the representation of an image into semantic and label latent variables, and simultaneously infer them in an intertwined fashion. To induce task-awareness, as part of the inference mechanics of TRIDENT, we exploit information across both query and support images of a few-shot task using a novel built-in attention-based transductive feature extraction module (we call AttFEX). Our extensive experimental results corroborate the efficacy of TRIDENT and demonstrate that, using the simplest of backbones, it sets a new state-of-the-art in the most commonly adopted datasets miniImageNet and tieredImageNet (offering up to 4% and 5% improvements, respectively), as well as for the recent challenging cross-domain miniImagenet --> CUB scenario offering a significant margin (up to 20% improvement) beyond the best existing cross-domain baselines. Code and experimentation can be found in our GitHub repository: this https URL

68.CADOps-Net: Jointly Learning CAD Operation Types and Steps from Boundary-Representations ⬇️

3D reverse engineering is a long sought-after, yet not completely achieved goal in the Computer-Aided Design (CAD) industry. The objective is to recover the construction history of a CAD model. Starting from a Boundary Representation (B-Rep) of a CAD model, this paper proposes a new deep neural network, CADOps-Net, that jointly learns the CAD operation types and the decomposition into different CAD operation steps. This joint learning allows to divide a B-Rep into parts that were created by various types of CAD operations at the same construction step; therefore providing relevant information for further recovery of the design history. Furthermore, we propose the novel CC3D-Ops dataset that includes over $37k$ CAD models annotated with CAD operation type labels and step labels. Compared to existing datasets, the complexity and variety of CC3D-Ops models are closer to those used for industrial purposes. Our experiments, conducted on the proposed CC3D-Ops and the publicly available Fusion360 datasets, demonstrate the competitive performance of CADOps-Net with respect to state-of-the-art, and confirm the importance of the joint learning of CAD operation types and steps.

69.Split-U-Net: Preventing Data Leakage in Split Learning for Collaborative Multi-Modal Brain Tumor Segmentation ⬇️

Split learning (SL) has been proposed to train deep learning models in a decentralized manner. For decentralized healthcare applications with vertical data partitioning, SL can be beneficial as it allows institutes with complementary features or images for a shared set of patients to jointly develop more robust and generalizable models. In this work, we propose "Split-U-Net" and successfully apply SL for collaborative biomedical image segmentation. Nonetheless, SL requires the exchanging of intermediate activation maps and gradients to allow training models across different feature spaces, which might leak data and raise privacy concerns. Therefore, we also quantify the amount of data leakage in common SL scenarios for biomedical image segmentation and provide ways to counteract such leakage by applying appropriate defense strategies.

70.InstanceFormer: An Online Video Instance Segmentation Framework ⬇️

Recent transformer-based offline video instance segmentation (VIS) approaches achieve encouraging results and significantly outperform online approaches. However, their reliance on the whole video and the immense computational complexity caused by full Spatio-temporal attention limit them in real-life applications such as processing lengthy videos. In this paper, we propose a single-stage transformer-based efficient online VIS framework named InstanceFormer, which is especially suitable for long and challenging videos. We propose three novel components to model short-term and long-term dependency and temporal coherence. First, we propagate the representation, location, and semantic information of prior instances to model short-term changes. Second, we propose a novel memory cross-attention in the decoder, which allows the network to look into earlier instances within a certain temporal window. Finally, we employ a temporal contrastive loss to impose coherence in the representation of an instance across all frames. Memory attention and temporal coherence are particularly beneficial to long-range dependency modeling, including challenging scenarios like occlusion. The proposed InstanceFormer outperforms previous online benchmark methods by a large margin across multiple datasets. Most importantly, InstanceFormer surpasses offline approaches for challenging and long datasets such as YouTube-VIS-2021 and OVIS. Code is available at this https URL.

71.The Value of AI Guidance in Human Examination of Synthetically-Generated Faces ⬇️

Face image synthesis has progressed beyond the point at which humans can effectively distinguish authentic faces from synthetically generated ones. Recently developed synthetic face image detectors boast "better-than-human" discriminative ability, especially those guided by human perceptual intelligence during the model's training process. In this paper, we investigate whether these human-guided synthetic face detectors can assist non-expert human operators in the task of synthetic image detection when compared to models trained without human-guidance. We conducted a large-scale experiment with more than 1,560 subjects classifying whether an image shows an authentic or synthetically-generated face, and annotate regions that supported their decisions. In total, 56,015 annotations across 3,780 unique face images were collected. All subjects first examined samples without any AI support, followed by samples given (a) the AI's decision ("synthetic" or "authentic"), (b) class activation maps illustrating where the model deems salient for its decision, or (c) both the AI's decision and AI's saliency map. Synthetic faces were generated with six modern Generative Adversarial Networks. Interesting observations from this experiment include: (1) models trained with human-guidance offer better support to human examination of face images when compared to models trained traditionally using cross-entropy loss, (2) binary decisions presented to humans offers better support than saliency maps, (3) understanding the AI's accuracy helps humans to increase trust in a given model and thus increase their overall accuracy. This work demonstrates that although humans supported by machines achieve better-than-random accuracy of synthetic face detection, the ways of supplying humans with AI support and of building trust are key factors determining high effectiveness of the human-AI tandem.

72.Toward Better Target Representation for Source-Free and Black-Box Domain Adaptation ⬇️

Domain adaptation aims at aligning the labeled source domain and the unlabeled target domain, and most existing approaches assume the source data is accessible. Unfortunately, this paradigm raises concerns in data privacy and security. Recent studies try to dispel these concerns by the Source-Free setting, which adapts the source-trained model towards target domain without exposing the source data. However, the Source-Free paradigm is still at risk of data leakage due to adversarial attacks to the source model. Hence, the Black-Box setting is proposed, where only the outputs of source model can be utilized. In this paper, we address both the Source-Free adaptation and the Black-Box adaptation, proposing a novel method named better target representation from Frequency Mixup and Mutual Learning (FMML). Specifically, we introduce a new data augmentation technique as Frequency MixUp, which highlights task-relevant objects in the interpolations, thus enhancing class-consistency and linear behavior for target models. Moreover, we introduce a network regularization method called Mutual Learning to the domain adaptation problem. It transfers knowledge inside the target model via self-knowledge distillation and thus alleviates overfitting on the source domain by learning multi-scale target representations. Extensive experiments show that our method achieves state-of-the-art performance on several benchmark datasets under both settings.

73.Locally temporal-spatial pattern learning with graph attention mechanism for EEG-based emotion recognition ⬇️

Technique of emotion recognition enables computers to classify human affective states into discrete categories. However, the emotion may fluctuate instead of maintaining a stable state even within a short time interval. There is also a difficulty to take the full use of the EEG spatial distribution due to its 3-D topology structure. To tackle the above issues, we proposed a locally temporal-spatial pattern learning graph attention network (LTS-GAT) in the present study. In the LTS-GAT, a divide-and-conquer scheme was used to examine local information on temporal and spatial dimensions of EEG patterns based on the graph attention mechanism. A dynamical domain discriminator was added to improve the robustness against inter-individual variations of the EEG statistics to learn robust EEG feature representations across different participants. We evaluated the LTS-GAT on two public datasets for affective computing studies under individual-dependent and independent paradigms. The effectiveness of LTS-GAT model was demonstrated when compared to other existing mainstream methods. Moreover, visualization methods were used to illustrate the relations of different brain regions and emotion recognition. Meanwhile, the weights of different time segments were also visualized to investigate emotion sparsity problems.

74.Robot Active Neural Sensing and Planning in Unknown Cluttered Environments ⬇️

Active sensing and planning in unknown, cluttered environments is an open challenge for robots intending to provide home service, search and rescue, narrow-passage inspection, and medical assistance. Although many active sensing methods exist, they often consider open spaces, assume known settings, or mostly do not generalize to real-world scenarios. We present the active neural sensing approach that generates the kinematically feasible viewpoint sequences for the robot manipulator with an in-hand camera to gather the minimum number of observations needed to reconstruct the underlying environment. Our framework actively collects the visual RGBD observations, aggregates them into scene representation, and performs object shape inference to avoid unnecessary robot interactions with the environment. We train our approach on synthetic data with domain randomization and demonstrate its successful execution via sim-to-real transfer in reconstructing narrow, covered, real-world cabinet environments cluttered with unknown objects. The natural cabinet scenarios impose significant challenges for robot motion and scene reconstruction due to surrounding obstacles and low ambient lighting conditions. However, despite unfavorable settings, our method exhibits high performance compared to its baselines in terms of various environment reconstruction metrics, including planning speed, the number of viewpoints, and overall scene coverage.

75.Unsupervised Anomaly Localization with Structural Feature-Autoencoders ⬇️

Unsupervised Anomaly Detection has become a popular method to detect pathologies in medical images as it does not require supervision or labels for training. Most commonly, the anomaly detection model generates a "normal" version of an input image, and the pixel-wise $l^p$-difference of the two is used to localize anomalies. However, large residuals often occur due to imperfect reconstruction of the complex anatomical structures present in most medical images. This method also fails to detect anomalies that are not characterized by large intensity differences to the surrounding tissue. We propose to tackle this problem using a feature-mapping function that transforms the input intensity images into a space with multiple channels where anomalies can be detected along different discriminative feature maps extracted from the original image. We then train an Autoencoder model in this space using structural similarity loss that does not only consider differences in intensity but also in contrast and structure. Our method significantly increases performance on two medical data sets for brain MRI. Code and experiments are available at this https URL

76.ULISSE: A Tool for One-shot Sky Exploration and its Application to Active Galactic Nuclei Detection ⬇️

Modern sky surveys are producing ever larger amounts of observational data, which makes the application of classical approaches for the classification and analysis of objects challenging and time-consuming. However, this issue may be significantly mitigated by the application of automatic machine and deep learning methods. We propose ULISSE, a new deep learning tool that, starting from a single prototype object, is capable of identifying objects sharing the same morphological and photometric properties, and hence of creating a list of candidate sosia. In this work, we focus on applying our method to the detection of AGN candidates in a Sloan Digital Sky Survey galaxy sample, since the identification and classification of Active Galactic Nuclei (AGN) in the optical band still remains a challenging task in extragalactic astronomy. Intended for the initial exploration of large sky surveys, ULISSE directly uses features extracted from the ImageNet dataset to perform a similarity search. The method is capable of rapidly identifying a list of candidates, starting from only a single image of a given prototype, without the need for any time-consuming neural network training. Our experiments show ULISSE is able to identify AGN candidates based on a combination of host galaxy morphology, color and the presence of a central nuclear source, with a retrieval efficiency ranging from 21% to 65% (including composite sources) depending on the prototype, where the random guess baseline is 12%. We find ULISSE to be most effective in retrieving AGN in early-type host galaxies, as opposed to prototypes with spiral- or late-type properties. Based on the results described in this work, ULISSE can be a promising tool for selecting different types of astrophysical objects in current and future wide-field surveys (e.g. Euclid, LSST etc.) that target millions of sources every single night.

77.The Value of Out-of-Distribution Data ⬇️

More data helps us generalize to a task. But real datasets can contain out-of-distribution (OOD) data; this can come in the form of heterogeneity such as intra-class variability but also in the form of temporal shifts or concept drifts. We demonstrate a counter-intuitive phenomenon for such problems: generalization error of the task can be a non-monotonic function of the number of OOD samples; a small number of OOD samples can improve generalization but if the number of OOD samples is beyond a threshold, then the generalization error can deteriorate. We also show that if we know which samples are OOD, then using a weighted objective between the target and OOD samples ensures that the generalization error decreases monotonically. We demonstrate and analyze this issue using linear classifiers on synthetic datasets and medium-sized neural networks on CIFAR-10.

78.Improving Computed Tomography (CT) Reconstruction via 3D Shape Induction ⬇️

Chest computed tomography (CT) imaging adds valuable insight in the diagnosis and management of pulmonary infectious diseases, like tuberculosis (TB). However, due to the cost and resource limitations, only X-ray images may be available for initial diagnosis or follow up comparison imaging during treatment. Due to their projective nature, X-rays images may be more difficult to interpret by clinicians. The lack of publicly available paired X-ray and CT image datasets makes it challenging to train a 3D reconstruction model. In addition, Chest X-ray radiology may rely on different device modalities with varying image quality and there may be variation in underlying population disease spectrum that creates diversity in inputs. We propose shape induction, that is, learning the shape of 3D CT from X-ray without CT supervision, as a novel technique to incorporate realistic X-ray distributions during training of a reconstruction model. Our experiments demonstrate that this process improves both the perceptual quality of generated CT and the accuracy of down-stream classification of pulmonary infectious diseases.

79.Satellite Image Search in AgoraEO ⬇️

The growing operational capability of global Earth Observation (EO) creates new opportunities for data-driven approaches to understand and protect our planet. However, the current use of EO archives is very restricted due to the huge archive sizes and the limited exploration capabilities provided by EO platforms. To address this limitation, we have recently proposed MiLaN, a content-based image retrieval approach for fast similarity search in satellite image archives. MiLaN is a deep hashing network based on metric learning that encodes high-dimensional image features into compact binary hash codes. We use these codes as keys in a hash table to enable real-time nearest neighbor search and highly accurate retrieval. In this demonstration, we showcase the efficiency of MiLaN by integrating it with EarthQube, a browser and search engine within AgoraEO. EarthQube supports interactive visual exploration and Query-by-Example over satellite image repositories. Demo visitors will interact with EarthQube playing the role of different users that search images in a large-scale remote sensing archive by their semantic content and apply other filters.

80.Time-lapse image classification using a diffractive neural network ⬇️

Diffractive deep neural networks (D2NNs) define an all-optical computing framework comprised of spatially engineered passive surfaces that collectively process optical input information by modulating the amplitude and/or the phase of the propagating light. Diffractive optical networks complete their computational tasks at the speed of light propagation through a thin diffractive volume, without any external computing power while exploiting the massive parallelism of optics. Diffractive networks were demonstrated to achieve all-optical classification of objects and perform universal linear transformations. Here we demonstrate, for the first time, a "time-lapse" image classification scheme using a diffractive network, significantly advancing its classification accuracy and generalization performance on complex input objects by using the lateral movements of the input objects and/or the diffractive network, relative to each other. In a different context, such relative movements of the objects and/or the camera are routinely being used for image super-resolution applications; inspired by their success, we designed a time-lapse diffractive network to benefit from the complementary information content created by controlled or random lateral shifts. We numerically explored the design space and performance limits of time-lapse diffractive networks, revealing a blind testing accuracy of 62.03% on the optical classification of objects from the CIFAR-10 dataset. This constitutes the highest inference accuracy achieved so far using a single diffractive network on the CIFAR-10 dataset. Time-lapse diffractive networks will be broadly useful for the spatio-temporal analysis of input signals using all-optical processors.

81.Aging prediction using deep generative model toward the development of preventive medicine ⬇️

From birth to death, we all experience surprisingly ubiquitous changes over time due to aging. If we can predict aging in the digital domain, that is, the digital twin of the human body, we would be able to detect lesions in their very early stages, thereby enhancing the quality of life and extending the life span. We observed that none of the previously developed digital twins of the adult human body explicitly trained longitudinal conversion rules between volumetric medical images with deep generative models, potentially resulting in poor prediction performance of, for example, ventricular volumes. Here, we establish a new digital twin of an adult human body that adopts longitudinally acquired head computed tomography (CT) images for training, enabling prediction of future volumetric head CT images from a single present volumetric head CT image. We, for the first time, adopt one of the three-dimensional flow-based deep generative models to realize this sequential three-dimensional digital twin. We show that our digital twin outperforms the latest methods of prediction of ventricular volumes in relatively short terms.

82.Extending nnU-Net is all you need ⬇️

Semantic segmentation is one of the most popular research areas in medical image computing. Perhaps surprisingly, despite its conceptualization dating back to 2018, nnU-Net continues to provide competitive out-of-the-box solutions for a broad variety of segmentation problems and is regularly used as a development framework for challenge-winning algorithms. Here we use nnU-Net to participate in the AMOS2022 challenge, which comes with a unique set of tasks: not only is the dataset one of the largest ever created and boasts 15 target structures, but the competition also requires submitted solutions to handle both MRI and CT scans. Through careful modification of nnU-net's hyperparameters, the addition of residual connections in the encoder and the design of a custom postprocessing strategy, we were able to substantially improve upon the nnU-Net baseline. Our final ensemble achieves Dice scores of 90.13 for Task 1 (CT) and 89.06 for Task 2 (CT+MRI) in a 5-fold cross-validation on the provided training cases.

83.Retinal Structure Detection in OCTA Image via Voting-based Multi-task Learning ⬇️

Automated detection of retinal structures, such as retinal vessels (RV), the foveal avascular zone (FAZ), and retinal vascular junctions (RVJ), are of great importance for understanding diseases of the eye and clinical decision-making. In this paper, we propose a novel Voting-based Adaptive Feature Fusion multi-task network (VAFF-Net) for joint segmentation, detection, and classification of RV, FAZ, and RVJ in optical coherence tomography angiography (OCTA). A task-specific voting gate module is proposed to adaptively extract and fuse different features for specific tasks at two levels: features at different spatial positions from a single encoder, and features from multiple encoders. In particular, since the complexity of the microvasculature in OCTA images makes simultaneous precise localization and classification of retinal vascular junctions into bifurcation/crossing a challenging task, we specifically design a task head by combining the heatmap regression and grid classification. We take advantage of three different \textit{en face} angiograms from various retinal layers, rather than following existing methods that use only a single \textit{en face}. To facilitate further research, part of these datasets with the source code and evaluation benchmark have been released for public access:this https URL.

84.Semi-Automatic Labeling and Semantic Segmentation of Gram-Stained Microscopic Images from DIBaS Dataset ⬇️

In this paper, a semi-automatic annotation of bacteria genera and species from DIBaS dataset is implemented using clustering and thresholding algorithms. A Deep learning model is trained to achieve the semantic segmentation and classification of the bacteria species. Classification accuracy of 95% is achieved. Deep learning models find tremendous applications in biomedical image processing. Automatic segmentation of bacteria from gram-stained microscopic images is essential to diagnose respiratory and urinary tract infections, detect cancers, etc. Deep learning will aid the biologists to get reliable results in less time. Additionally, a lot of human intervention can be reduced. This work can be helpful to detect bacteria from urinary smear images, sputum smear images, etc to diagnose urinary tract infections, tuberculosis, pneumonia, etc.

85.Ultra-high-resolution unpaired stain transformation via Kernelized Instance Normalization ⬇️

While hematoxylin and eosin (H&E) is a standard staining procedure, immunohistochemistry (IHC) staining further serves as a diagnostic and prognostic method. However, acquiring special staining results requires substantial costs.
Hence, we proposed a strategy for ultra-high-resolution unpaired image-to-image translation: Kernelized Instance Normalization (KIN), which preserves local information and successfully achieves seamless stain transformation with constant GPU memory usage. Given a patch, corresponding position, and a kernel, KIN computes local statistics using convolution operation. In addition, KIN can be easily plugged into most currently developed frameworks without re-training.
We demonstrate that KIN achieves state-of-the-art stain transformation by replacing instance normalization (IN) layers with KIN layers in three popular frameworks and testing on two histopathological datasets. Furthermore, we manifest the generalizability of KIN with high-resolution natural images. Finally, human evaluation and several objective metrics are used to compare the performance of different approaches.
Overall, this is the first successful study for the ultra-high-resolution unpaired image-to-image translation with constant space complexity. Code is available at: this https URL

86.CM-MLP: Cascade Multi-scale MLP with Axial Context Relation Encoder for Edge Segmentation of Medical Image ⬇️

The convolutional-based methods provide good segmentation performance in the medical image segmentation task. However, those methods have the following challenges when dealing with the edges of the medical images: (1) Previous convolutional-based methods do not focus on the boundary relationship between foreground and background around the segmentation edge, which leads to the degradation of segmentation performance when the edge changes complexly. (2) The inductive bias of the convolutional layer cannot be adapted to complex edge changes and the aggregation of multiple-segmented areas, resulting in its performance improvement mostly limited to segmenting the body of segmented areas instead of the edge. To address these challenges, we propose the CM-MLP framework on MFI (Multi-scale Feature Interaction) block and ACRE (Axial Context Relation Encoder) block for accurate segmentation of the edge of medical image. In the MFI block, we propose the cascade multi-scale MLP (Cascade MLP) to process all local information from the deeper layers of the network simultaneously and utilize a cascade multi-scale mechanism to fuse discrete local information gradually. Then, the ACRE block is used to make the deep supervision focus on exploring the boundary relationship between foreground and background to modify the edge of the medical image. The segmentation accuracy (Dice) of our proposed CM-MLP framework reaches 96.96%, 96.76%, and 82.54% on three benchmark datasets: CVC-ClinicDB dataset, sub-Kvasir dataset, and our in-house dataset, respectively, which significantly outperform the state-of-the-art method. The source code and trained models will be available at this https URL.

87.Three-dimensional micro-structurally informed in silico myocardium -- towards virtual imaging trials in cardiac diffusion weighted MRI ⬇️

In silico tissue models enable evaluating quantitative models of magnetic resonance imaging. This includes validating and sensitivity analysis of imaging biomarkers and tissue microstructure parameters. We propose a novel method to generate a realistic numerical phantom of myocardial microstructure. We extend previous studies accounting for the cardiomyocyte shape variability, water exchange between the cardiomyocytes (intercalated discs), myocardial microstructure disarray, and four sheetlet orientations. In the first stage of the method, cardiomyocytes and sheetlets are generated by considering the shape variability and intercalated discs in cardiomyocyte-to-cardiomyocyte connections. Sheetlets are then aggregated and oriented in the directions of interest. Our morphometric study demonstrates no significant difference ($p>0.01$) between the distribution of volume, length, and primary and secondary axes of the numerical and real (literature) cardiomyocyte data. Structural correlation analysis validates that the in-silico tissue is in the same class of disorderliness as the real tissue. Additionally, the absolute angle differences between the simulated helical angle (HA) and input HA (reference value) of the cardiomyocytes ($4.3^\circ\pm 3.1^\circ$) demonstrate a good agreement with the absolute angle difference between the measured HA using experimental cardiac diffusion tensor imaging (cDTI) and histology (reference value) reported by (Holmes et al., 2000) ($3.7^\circ\pm6.4^\circ$) and (Scollan et al., 1998) ($4.9^\circ\pm 14.6^\circ$). The angular distance between eigenvectors and sheetlet angles of the input and simulated cDTI is smaller than those between measured angles using structural tensor imaging (gold standard) and experimental cDTI. These results confirm that the proposed method can generate richer numerical phantoms for the myocardium than previous studies.

88.RIBAC: Towards Robust and Imperceptible Backdoor Attack against Compact DNN ⬇️

Recently backdoor attack has become an emerging threat to the security of deep neural network (DNN) models. To date, most of the existing studies focus on backdoor attack against the uncompressed model; while the vulnerability of compressed DNNs, which are widely used in the practical applications, is little exploited yet. In this paper, we propose to study and develop Robust and Imperceptible Backdoor Attack against Compact DNN models (RIBAC). By performing systematic analysis and exploration on the important design knobs, we propose a framework that can learn the proper trigger patterns, model parameters and pruning masks in an efficient way. Thereby achieving high trigger stealthiness, high attack success rate and high model efficiency simultaneously. Extensive evaluations across different datasets, including the test against the state-of-the-art defense mechanisms, demonstrate the high robustness, stealthiness and model efficiency of RIBAC. Code is available at this https URL

89.Estimation Contracts for Outlier-Robust Geometric Perception ⬇️

Outlier-robust estimation is a fundamental problem and has been extensively investigated by statisticians and practitioners. The last few years have seen a convergence across research fields towards "algorithmic robust statistics", which focuses on developing tractable outlier-robust techniques for high-dimensional estimation problems. Despite this convergence, research efforts across fields have been mostly disconnected from one another. This paper bridges recent work on certifiable outlier-robust estimation for geometric perception in robotics and computer vision with parallel work in robust statistics. In particular, we adapt and extend recent results on robust linear regressions (applicable to the low-outlier case with << 50% outliers) and list-decodable regression (applicable to the high-outlier case with >> 50% outliers) to the setup commonly found in robotics and vision, where (i) variables (e.g., rotations, poses) belong to a non-convex domain, (ii) measurements are vector-valued, and (iii) the number of outliers is not known a priori. The emphasis here is on performance guarantees: rather than proposing new algorithms, we provide conditions on the input measurements under which modern estimation algorithms are guaranteed to recover an estimate close to the ground truth in the presence of outliers. These conditions are what we call an "estimation contract". Besides the proposed extensions of existing results, we believe the main contributions of this paper are (i) to unify parallel research lines by pointing out commonalities and differences, (ii) to introduce advanced material (e.g., sum-of-squares proofs) in an accessible and self-contained presentation for the practitioner, and (iii) to point out a few immediate opportunities and open questions in outlier-robust geometric perception.

90.Memristive Computing for Efficient Inference on Resource Constrained Devices ⬇️

The advent of deep learning has resulted in a number of applications which have transformed the landscape of the research area in which it has been applied. However, with an increase in popularity, the complexity of classical deep neural networks has increased over the years. As a result, this has leads to considerable problems during deployment on devices with space and time constraints. In this work, we perform a review of the present advancements in non-volatile memory and how the use of resistive RAM memory, particularly memristors, can help to progress the state of research in deep learning. In other words, we wish to present an ideology that advances in the field of memristive technology can greatly influence and impact deep learning inference on edge devices.