1.Unsupervised Learning of Shape and Pose with Differentiable Point Clouds pdf
We address the problem of learning accurate 3D shape and camera pose from a collection of unlabeled category-specific images. We train a convolutional network to predict both the shape and the pose from a single image by minimizing the reprojection error: given several views of an object, the projections of the predicted shapes to the predicted camera poses should match the provided views. To deal with pose ambiguity, we introduce an ensemble of pose predictors which we then distill to a single "student" model. To allow for efficient learning of high-fidelity shapes, we represent the shapes by point clouds and devise a formulation allowing for differentiable projection of these. Our experiments show that the distilled ensemble of pose predictors learns to estimate the pose accurately, while the point cloud representation allows to predict detailed shape models. The supplementary video can be found at this https URL
2.Brain Tumor Image Retrieval via Multitask Learning pdf
Classification-based image retrieval systems are built by training convolutional neural networks (CNNs) on a relevant classification problem and using the distance in the resulting feature space as a similarity metric. However, in practical applications, it is often desirable to have representations which take into account several aspects of the data (e.g., brain tumor type and its localization). In our work, we extend the classification-based approach with multitask learning: we train a CNN on brain MRI scans with heterogeneous labels and implement a corresponding tumor image retrieval system. We validate our approach on brain tumor data which contains information about tumor types, shapes and localization. We show that our method allows us to build representations that contain more relevant information about tumors than single-task classification-based approaches.
3.Generation of Virtual Dual Energy Images from Standard Single-Shot Radiographs using Multi-scale and Conditional Adversarial Network pdf
Dual-energy (DE) chest radiographs provide greater diagnostic information than standard radiographs by separating the image into bone and soft tissue, revealing suspicious lesions which may otherwise be obstructed from view. However, acquisition of DE images requires two physical scans, necessitating specialized hardware and processing, and images are prone to motion artifact. Generation of virtual DE images from standard, single-shot chest radiographs would expand the diagnostic value of standard radiographs without changing the acquisition procedure. We present a Multi-scale Conditional Adversarial Network (MCA-Net) which produces high-resolution virtual DE bone images from standard, single-shot chest radiographs. Our proposed MCA-Net is trained using the adversarial network so that it learns sharp details for the production of high-quality bone images. Then, the virtual DE soft tissue image is generated by processing the standard radiograph with the virtual bone image using a cross projection transformation. Experimental results from 210 patient DE chest radiographs demonstrated that the algorithm can produce high-quality virtual DE chest radiographs. Important structures were preserved, such as coronary calcium in bone images and lung lesions in soft tissue images. The average structure similarity index and the peak signal to noise ratio of the produced bone images in testing data were 96.4 and 41.5, which are significantly better than results from previous methods. Furthermore, our clinical evaluation results performed on the publicly available dataset indicates the clinical values of our algorithms. Thus, our algorithm can produce high-quality DE images that are potentially useful for radiologists, computer-aided diagnostics, and other diagnostic tasks.
4.Baseline Detection in Historical Documents using Convolutional U-Nets pdf
Baseline detection is still a challenging task for heterogeneous collections of historical documents. We present a novel approach to baseline extraction in such settings, turning out the winning entry to the ICDAR 2017 Competition on Baseline detection (cBAD). It utilizes deep convolutional nets (CNNs) for both, the actual extraction of baselines, as well as for a simple form of layout analysis in a pre-processing step. To the best of our knowledge it is the first CNN-based system for baseline extraction applying a U-net architecture and sliding window detection, profiting from a high local accuracy of the candidate lines extracted. Final baseline post-processing complements our approach, compensating for inaccuracies mainly due to missing context information during sliding window detection. We experimentally evaluate the components of our system individually on the cBAD dataset. Moreover, we investigate how it generalizes to different data by means of the dataset used for the baseline extraction task of the ICDAR 2017 Competition on Layout Analysis for Challenging Medieval Manuscripts (HisDoc). A comparison with the results reported for HisDoc shows that it also outperforms the contestants of the latter.
5.Improving Annotation for 3D Pose Dataset of Fine-Grained Object Categories pdf
Existing 3D pose datasets of object categories are limited to generic object types and lack of fine-grained information. In this work, we introduce a new large-scale dataset that consists of 409 fine-grained categories and 31,881 images with accurate 3D pose annotation. Specifically, we augment three existing fine-grained object recognition datasets (StanfordCars, CompCars and FGVC-Aircraft) by finding a specific 3D model for each sub-category from ShapeNet and manually annotating each 2D image by adjusting a full set of 7 continuous perspective parameters. Since the fine-grained shapes allow 3D models to better fit the images, we further improve the annotation quality by initializing from the human annotation and conducting local search of the pose parameters with the objective of maximizing the IoUs between the projected mask and the segmentation reference estimated from state-of-the-art deep Convolutional Neural Networks (CNNs). We provide full statistics of the annotations with qualitative and quantitative comparisons suggesting that our dataset can be a complementary source for studying 3D pose estimation. The dataset can be downloaded at this http URL.
6.Field Of Interest Proposal for Augmented Mitotic Cell Count: Comparison of two Convolutional Networks pdf
Most tumor grading systems for human as for veterinary histopathology are based upon the absolute count of mitotic figures in a certain reference area of a histology slide. Since time for prognostication is limited in a diagnostic setting, the pathologist will often almost arbitrarily choose a certain field of interest assumed to have the highest mitotic activity. However, as mitotic figures are commonly very sparse on the slide and often have a patchy distribution, this poses a sampling problem which is known to be able to influence the tumor prognostication. On the other hand, automatic detection of mitotic figures can't yet be considered reliable enough for clinical application. In order to aid the work of the human expert and at the same time reduce variance in tumor grading, it is beneficial to assess the whole slide image (WSI) for the highest mitotic activity and use this as a reference region for human counting. For this task, we compare two methods for region of interest proposal, both based on convolutional neural networks (CNN). For both approaches, the CNN performs a segmentation of the WSI to assess mitotic activity. The first method performs a segmentation at the original image resolution, while the second approach performs a segmentation operation at a significantly reduced resolution, cutting down on processing complexity. We evaluate the approach using a dataset of 32 completely annotated whole slide images of canine mast cell tumors, where 22 were used for training of the network and 10 for test. Our results indicate that, while the overall correlation to the ground truth mitotic activity is considerably higher (0.94 vs. 0.83) for the approach based upon the fine resolution network, the field of interest choices are only marginally better. Both approaches propose fields of interest that contain a mitotic count in the upper quartile of respective slides.
7.Dating Ancient Paintings of Mogao Grottoes Using Deeply Learnt Visual Codes pdf
Cultural heritage is the asset of all the peoples of the world. The preservation and inheritance of cultural heritage is conducive to the progress of human civilization. In northwestern China, there is a world heritage site -- Mogao Grottoes -- that has a plenty of mural paintings showing the historical cultures of ancient China. To study these historical cultures, one critical procedure is to date the mural paintings, i.e., determining the era when they were created. Until now, most mural paintings at Mogao Grottoes have been dated by directly referring to the mural texts or historical documents. However, some are still left with creation-era undetermined due to the lack of reference materials. Considering that the drawing style of mural paintings was changing along the history and the drawing style can be learned and quantified through painting data, we formulate the problem of mural-painting dating into a problem of drawing-style classification. In fact, drawing styles can be expressed not only in color or curvature, but also in some unknown forms -- the forms that have not been observed. To this end, besides sophisticated color and shape descriptors, a deep convolution neural network is designed to encode the implicit drawing styles. 3860 mural paintings collected from 194 different grottoes with determined creation-era labels are used to train the classification model and build the dating method. In experiments, the proposed dating method is applied to seven mural paintings which were previously dated with controversies, and the exciting new dating results are approved by the Dunhuang expert.
8.Exploring Correlations in Multiple Facial Attributes through Graph Attention Network pdf
Estimating multiple attributes from a single facial image gives comprehensive descriptions on the high level semantics of the face. It is naturally regarded as a multi-task supervised learning problem with a single deep CNN, in which lower layers are shared, and higher ones are task-dependent with the multi-branch structure. Within the traditional deep multi-task learning (DMTL) framework, this paper intends to fully exploit the correlations among different attributes by constructing a graph. The node in graph represents the feature vector from a particular branch for a given attribute, and the edge can be defined by either the prior knowledge or the similarity between two nodes in the embedding with a fully data-driven manner. We analyze that the attention mechanism actually takes effect in the latter case, and utilize the Graph Attention Layer (GAL) for exploring on the most relevant attribute feature and refining the task-dependant feature by considering other attributes. Experiments show that by mining the correlations among attributes, our method can improve the recognition accuracy on CelebA and LFWA dataset. And it also achieves competitive performance.
9.Atrial scars segmentation via potential learning in the graph-cuts framework pdf
Late Gadolinium Enhancement Magnetic Resonance Imaging (LGE MRI) emerged as a routine scan for patients with atrial fibrillation (AF). However, due to the low image quality automating the quantification and analysis of the atrial scars is challenging. In this study, we pro-posed a fully automated method based on the graph-cuts framework, where the potential of the graph is learned on a surface mesh of the left atrium (LA) using an equidistant projection and a Deep Neural Network (DNN). For validation, we employed 100 datasets with manual delineation. The results showed that the performance of the proposed method improved and converged with respect to the increased size of training patches, which provide important features of the structural and texture information learned by the DNN. The segmentation could be further improved when the contribution from the t-link and n-link is balanced, thanks to inter-relationship learned by the DNN for the graph-cuts algorithm. Compared with the published methods which mostly acquired manual delineation of the LA or LA wall, our method is fully automatic and demonstrated evidently better results with statistical significance. Finally, the accuracy of quantifying the scars assessed by the Dice score was 0.570. The results are promising and the method can be useful in diagnosis and prognosis of AF.
10.Learning to Measure Change: Fully Convolutional Siamese Metric Networks for Scene Change Detection pdf
The key factor of scene change detection is to learn effective feature to have higher similarity for changed parts and lower similarity for unchanged parts. However, existed researches on change detection are based on semantic segmentation, which gives rise to the lack of semantic discriminative in feature space. In this paper, we propose a novel Fully Convolutional siamese metric Network (CosimNet) and utilize various loss functions, including original contrastive loss and threshold hinge loss to address the limitation above, aiming to pull together similar pairs and push apart dissimilar pairs in the feature space. We demonstrate the effectiveness of the proposed approach with experiments on three challenging datasets including CDnet, PCD2015, and VL-CMU. Source code is available at this https URL
11.SG-One: Similarity Guidance Network for One-Shot Semantic Segmentation pdf
One-shot semantic segmentation poses a challenging task of recognizing the object regions from unseen categories with only one annotated example as supervision. In this paper, we propose a simple yet effective Similarity Guidance network to tackle the One-shot (SG-One) segmentation problem. We aim at predicting the segmentation mask of a query image with the reference to one densely labeled support image. To obtain the robust representative feature of the support image, we firstly propose a masked average pooling strategy for producing the guidance features using only the pixels belonging to the support image. We then leverage the cosine similarity to build the relationship between the guidance features and features of pixels from the query image. In this way, the possibilities embedded in the produced similarity maps can be adopted to guide the process of segmenting objects. Furthermore, our SG-One is a unified framework which can efficiently process both support and query images within one network and be learned in an end-to-end manner. We conduct extensive experiments on Pascal VOC 2012. In particular, our SG-One achieves the mIoU score of 46.3%, which outperforms the state-of-the-art.
12.Atrial fibrosis quantification based on maximum likelihood estimator of multivariate images pdf
We present a fully-automated segmentation and quantification of the left atrial (LA) fibrosis and scars combining two cardiac MRIs, one is the target late gadolinium-enhanced (LGE) image, and the other is an anatomical MRI from the same acquisition session. We formulate the joint distribution of images using a multivariate mixture model (MvMM), and employ the maximum likelihood estimator (MLE) for texture classification of the images simultaneously. The MvMM can also embed transformations assigned to the images to correct the misregistration. The iterated conditional mode algorithm is adopted for optimization. This method first extracts the anatomical shape of the LA, and then estimates a prior probability map. It projects the resulting segmentation onto the LA surface, for quantification and analysis of scarring. We applied the proposed method to 36 clinical data sets and obtained promising results (Accuracy:
$0.809\pm .150$ , Dice:$0.556\pm.187$ ). We compared the method with the conventional algorithms and showed an evidently and statistically better performance ($p<0.03$ ).
13.Where is this? Video geolocation based on neural network features pdf
In this work we propose a method that geolocates videos within a delimited widespread area based solely on the frames visual content. Our proposed method tackles video-geolocation through traditional image retrieval techniques considering Google Street View as the reference point. To achieve this goal we use the deep learning features obtained from NetVLAD to represent images, since through this feature vectors the similarity is their L2 norm. In this paper, we propose a family of voting-based methods to aggregate frame-wise geolocation results which boost the video geolocation result. The best aggregation found through our experiments considers both NetVLAD and SIFT similarity, as well as the geolocation density of the most similar results. To test our proposed method, we gathered a new video dataset from Pittsburgh Downtown area to benefit and stimulate more work in this area. Our system achieved a precision of 90% while geolocating videos within a range of 150 meters or two blocks away from the original position.
14.VIENA2: A Driving Anticipation Dataset pdf
Action anticipation is critical in scenarios where one needs to react before the action is finalized. This is, for instance, the case in automated driving, where a car needs to, e.g., avoid hitting pedestrians and respect traffic lights. While solutions have been proposed to tackle subsets of the driving anticipation tasks, by making use of diverse, task-specific sensors, there is no single dataset or framework that addresses them all in a consistent manner. In this paper, we therefore introduce a new, large-scale dataset, called VIENA2, covering 5 generic driving scenarios, with a total of 25 distinct action classes. It contains more than 15K full HD, 5s long videos acquired in various driving conditions, weathers, daytimes and environments, complemented with a common and realistic set of sensor measurements. This amounts to more than 2.25M frames, each annotated with an action label, corresponding to 600 samples per action class. We discuss our data acquisition strategy and the statistics of our dataset, and benchmark state-of-the-art action anticipation techniques, including a new multi-modal LSTM architecture with an effective loss function for action anticipation in driving scenarios.
15.Hierarchical ResNeXt Models for Breast Cancer Histology Image Classification pdf
Microscopic histology image analysis is a cornerstone in early detection of breast cancer. However these images are very large and manual analysis is error prone and very time consuming. Thus automating this process is in high demand. We proposed a hierarchical system of convolutional neural networks (CNN) that classifies automatically patches of these images into four pathologies: normal, benign, in situ carcinoma and invasive carcinoma. We evaluated our system on the BACH challenge dataset of image-wise classification and a small dataset that we used to extend it. Using a train/test split of 75%/25%, we achieved an accuracy rate of 0.99 on the test split for the BACH dataset and 0.96 on that of the extension. On the test of the BACH challenge, we've reached an accuracy of 0.81 which rank us to the 8th out of 51 teams.
16.C2A: Crowd Consensus Analytics for Virtual Colonoscopy pdf
We present a medical crowdsourcing visual analytics platform called C{$^2$}A to visualize, classify and filter crowdsourced clinical data. More specifically, C$^2$A is used to build consensus on a clinical diagnosis by visualizing crowd responses and filtering out anomalous activity. Crowdsourcing medical applications have recently shown promise where the non-expert users (the crowd) were able to achieve accuracy similar to the medical experts. This has the potential to reduce interpretation/reading time and possibly improve accuracy by building a consensus on the findings beforehand and letting the medical experts make the final diagnosis. In this paper, we focus on a virtual colonoscopy (VC) application with the clinical technicians as our target users, and the radiologists acting as consultants and classifying segments as benign or malignant. In particular, C$^2$A is used to analyze and explore crowd responses on video segments, created from fly-throughs in the virtual colon. C$^2$A provides several interactive visualization components to build crowd consensus on video segments, to detect anomalies in the crowd data and in the VC video segments, and finally, to improve the non-expert user's work quality and performance by A/B testing for the optimal crowdsourcing platform and application-specific parameters. Case studies and domain experts feedback demonstrate the effectiveness of our framework in improving workers' output quality, the potential to reduce the radiologists' interpretation time, and hence, the potential to improve the traditional clinical workflow by marking the majority of the video segments as benign based on the crowd consensus.
17.3D shape retrieval basing on representatives of classes pdf
In this paper, we present an improvement of our proposed technique for 3D shape retrieval in classified databases [2] which is based on representatives of classes. Instead of systematically matching the object-query with all 3D models of the database, our idea presented in [2] consist, for a classified database, to represent each class by one representative that is used to orient the retrieval process to the right class (the class excepted to contain 3D models similar to the query). In order to increase the chance to fall in the right class, our idea in this work is to represent each class by more than one representative. In this case, instead of using only one representative to decide which is the right class we use a set of representatives this will contribute certainly to improving the relevance of retrieval results. The obtained experimental results show that the relevance is significantly improved.
18.Visualization Framework for Colonoscopy Videos pdf
We present a visualization framework for annotating and comparing colonoscopy videos, where these annotations can then be used for semi-automatic report generation at the end of the procedure. Currently, there are approximately 14 million colonoscopies performed every year in the US. In this work, we create a visualization tool to deal with the deluge of colonoscopy videos in a more effective way. We present an interactive visualization framework for the annotation and tagging of colonoscopy videos in an easy and intuitive way. These annotations and tags can later be used for report generation for electronic medical records and for comparison at an individual as well as group level. We also present important use cases and medical expert feedback for our visualization framework.
19.Learning Spectral Transform Network on 3D Surface for Non-rigid Shape Analysis pdf
Designing a network on 3D surface for non-rigid shape analysis is a challenging task. In this work, we propose a novel spectral transform network on 3D surface to learn shape descriptors. The proposed network architecture consists of four stages: raw descriptor extraction, surface second-order pooling, mixture of power function-based spectral transform, and metric learning. The proposed network is simple and shallow. Quantitative experiments on challenging benchmarks show its effectiveness for non-rigid shape retrieval and classification, e.g., it achieved the highest accuracies on SHREC14, 15 datasets as well as the Range subset of SHREC17 dataset.
20.Electricity consumption forecasting method based on MPSO-BP neural network model pdf
This paper deals with the problem of the electricity consumption forecasting method. An MPSO-BP (modified particle swarm optimization-back propagation) neural network model is constructed based on the history data of a mineral company of Anshan in China. The simulation showed that the convergence of the algorithm and forecasting accuracy using the obtained model are better than those of other traditional ones, such as BP, PSO, fuzzy neural network and so on. Then we predict the electricity consumption of each month in 2017 based on the MPSO-BP neural network model.
21.Automated identification of hookahs (waterpipes) on Instagram: an application in feature extraction using Convolutional Neural Network and Support Vector Machine classification pdf
Background: Instagram, with millions of posts per day, can be used to inform public health surveillance targets and policies. However, current research relying on image-based data often relies on hand coding of images which is time consuming and costly, ultimately limiting the scope of the study. Current best practices in automated image classification (e.g., support vector machine (SVM), Backpropagation (BP) neural network, and artificial neural network) are limited in their capacity to accurately distinguish between objects within images. Objective: This study demonstrates how convolutional neural network (CNN) can be used to extract unique features within an image and how SVM can then be used to classify the image. Methods: Images of waterpipes or hookah (an emerging tobacco product possessing similar harms to that of cigarettes) were collected from Instagram and used in analyses (n=840). CNN was used to extract unique features from images identified to contain waterpipes. A SVM classifier was built to distinguish between images with and without waterpipes. Methods for image classification were then compared to show how a CNN + SVM classifier could improve accuracy. Results: As the number of the validated training images increased, the total number of extracted features increased. Additionally, as the number of features learned by the SVM classifier increased, the average level of accuracy increased. Overall, 99.5% of the 420 images classified were correctly identified as either hookah or non-hookah images. This level of accuracy was an improvement over earlier methods that used SVM, CNN or Bag of Features (BOF) alone. Conclusions: CNN extracts more features of the images allowing a SVM classifier to be better informed, resulting in higher accuracy compared with methods that extract fewer features. Future research can use this method to grow the scope of image-based studies.
22.A Regressive Convolution Neural network and Support Vector Regression Model for Electricity Consumption Forecasting pdf
Electricity consumption forecasting has important implications for the mineral companies on guiding quarterly work, normal power system operation, and the management. However, electricity consumption prediction for the mineral company is different from traditional electricity load prediction since mineral company electricity consumption can be affected by various factors (e.g., ore grade, processing quantity of the crude ore, ball milling fill rate). The problem is non-trivial due to three major challenges for traditional methods: insufficient training data, high computational cost and low prediction accu-racy. To tackle these challenges, we firstly propose a Regressive Convolution Neural Network (RCNN) to predict the electricity consumption. While RCNN still suffers from high computation overhead, we utilize RCNN to extract features from the history data and Regressive Support Vector Machine (SVR) trained with the features to predict the electricity consumption. The experimental results show that the proposed RCNN-SVR model achieves higher accuracy than using the traditional RNN or SVM alone. The MSE, MAPE, and CV-RMSE of RCNN-SVR model are 0.8564, 1.975%, and 0.0687% respectively, which illustrates the low predicting error rate of the proposed model.
23.Corresponding Supine and Prone Colon Visualization Using Eigenfunction Analysis and Fold Modeling pdf
We present a method for registration and visualization of corresponding supine and prone virtual colonoscopy scans based on eigenfunction analysis and fold modeling. In virtual colonoscopy, CT scans are acquired with the patient in two positions, and their registration is desirable so that physicians can corroborate findings between scans. Our algorithm performs this registration efficiently through the use of Fiedler vector representation (the second eigenfunction of the Laplace-Beltrami operator). This representation is employed to first perform global registration of the two colon positions. The registration is then locally refined using the haustral folds, which are automatically segmented using the 3D level sets of the Fiedler vector. The use of Fiedler vectors and the segmented folds presents a precise way of visualizing corresponding regions across datasets and visual modalities. We present multiple methods of visualizing the results, including 2D flattened rendering and the corresponding 3D endoluminal views. The precise fold modeling is used to automatically find a suitable cut for the 2D flattening, which provides a less distorted visualization. Our approach is robust, and we demonstrate its efficiency and efficacy by showing matched views on both the 2D flattened colons and in the 3D endoluminal view. We analytically evaluate the results by measuring the distance between features on the registered colons, and we also assess our fold segmentation against 20 manually labeled datasets. We have compared our results analytically to previous methods, and have found our method to achieve superior results. We also prove the hot spots conjecture for modeling cylindrical topology using Fiedler vector representation, which allows our approach to be used for general cylindrical geometry modeling and feature extraction.
24.Improved Techniques for GAN based Facial Inpainting pdf
In this paper we present several architectural and optimization recipes for generative adversarial network(GAN) based facial semantic inpainting. Current benchmark models are susceptible to initial solutions of non-convex optimization criterion of GAN based inpainting. We present an end-to-end trainable parametric network to deterministically start from good initial solutions leading to more photo realistic reconstructions with significant optimization speed up. For the first time, we show how to efficiently extend GAN based single image inpainter models to sequences by a)learning to initialize a temporal window of solutions with a recurrent neural network and b)imposing a temporal smoothness loss(during iterative optimization) to respect the redundancy in temporal dimension of a sequence. We conduct comprehensive empirical evaluations on CelebA images and pseudo sequences followed by real life videos of VidTIMIT dataset. The proposed method significantly outperforms current GAN based state-of-the-art in terms of reconstruction quality with a simultaneous speedup of over 15$\times$. We also show that our proposed model is better in preserving facial identity in a sequence even without explicitly using any face recognition module during training.
25.Image Inpainting via Generative Multi-column Convolutional Neural Networks pdf
In this paper, we propose a generative multi-column network for image inpainting. This network synthesizes different image components in a parallel manner within one stage. To better characterize global structures, we design a confidence-driven reconstruction loss while an implicit diversified MRF regularization is adopted to enhance local details. The multi-column network combined with the reconstruction and MRF loss propagates local and global information derived from context to the target inpainting regions. Extensive experiments on challenging street view, face, natural objects and scenes manifest that our method produces visual compelling results even without previously common post-processing.
26.Sequential Context Encoding for Duplicate Removal pdf
Duplicate removal is a critical step to accomplish a reasonable amount of predictions in prevalent proposal-based object detection frameworks. Albeit simple and effective, most previous algorithms utilize a greedy process without making sufficient use of properties of input data. In this work, we design a new two-stage framework to effectively select the appropriate proposal candidate for each object. The first stage suppresses most of easy negative object proposals, while the second stage selects true positives in the reduced proposal set. These two stages share the same network structure, \ie, an encoder and a decoder formed as recurrent neural networks (RNN) with global attention and context gate. The encoder scans proposal candidates in a sequential manner to capture the global context information, which is then fed to the decoder to extract optimal proposals. In our extensive experiments, the proposed method outperforms other alternatives by a large margin.
27.MEMC-Net: Motion Estimation and Motion Compensation Driven Neural Network for Video Interpolation and Enhancement pdf
Motion estimation (ME) and motion compensation (MC) have been widely used for classical video frame interpolation systems over the past decades. Recently, a number of data-driven frame interpolation methods based on convolutional neural networks have been proposed. However, existing learning based methods typically estimate either flow or compensation kernels, thereby limiting performance on both computational efficiency and interpolation accuracy. In this work, we propose a motion estimation and compensation driven neural network for video frame interpolation. A novel adaptive warping layer is developed to integrate both optical flow and interpolation kernels to synthesize target frame pixels. This layer is fully differentiable such that both the flow and kernel estimation networks can be optimized jointly. The proposed model benefits from the advantages of motion estimation and compensation methods without using hand-crafted features. Compared to existing methods, our approach is computationally efficient and able to generate more visually appealing results. Furthermore, the proposed MEMC-Net can be seamlessly adapted to several video enhancement tasks, e.g., super-resolution, denoising, and deblocking. Extensive quantitative and qualitative evaluations demonstrate that the proposed method performs favorably against the state-of-the-art video frame interpolation and enhancement algorithms on a wide range of datasets.
28.Left Ventricle Segmentation via Optical-Flow-Net from Short-axis Cine MRI: Preserving the Temporal Coherence of Cardiac Motion pdf
Quantitative assessment of left ventricle (LV) function from cine MRI has significant diagnostic and prognostic value for cardiovascular disease patients. The temporal movement of LV provides essential information on the contracting/relaxing pattern of heart, which is keenly evaluated by clinical experts in clinical practice. Inspired by the expert way of viewing Cine MRI, we propose a new CNN module that is able to incorporate the temporal information into LV segmentation from cine MRI. In the proposed CNN, the optical flow (OF) between neighboring frames is integrated and aggregated at feature level, such that temporal coherence in cardiac motion can be taken into account during segmentation. The proposed module is integrated into the U-net architecture without need of additional training. Furthermore, dilated convolution is introduced to improve the spatial accuracy of segmentation. Trained and tested on the Cardiac Atlas database, the proposed network resulted in a Dice index of 95% and an average perpendicular distance of 0.9 pixels for the middle LV contour, significantly outperforming the original U-net that processes each frame individually. Notably, the proposed method improved the temporal coherence of LV segmentation results, especially at the LV apex and base where the cardiac motion is difficult to follow.
29.Learning Material-Aware Local Descriptors for 3D Shapes pdf
Material understanding is critical for design, geometric modeling, and analysis of functional objects. We enable material-aware 3D shape analysis by employing a projective convolutional neural network architecture to learn material- aware descriptors from view-based representations of 3D points for point-wise material classification or material- aware retrieval. Unfortunately, only a small fraction of shapes in 3D repositories are labeled with physical mate- rials, posing a challenge for learning methods. To address this challenge, we crowdsource a dataset of 3080 3D shapes with part-wise material labels. We focus on furniture models which exhibit interesting structure and material variabil- ity. In addition, we also contribute a high-quality expert- labeled benchmark of 115 shapes from Herman-Miller and IKEA for evaluation. We further apply a mesh-aware con- ditional random field, which incorporates rotational and reflective symmetries, to smooth our local material predic- tions across neighboring surface patches. We demonstrate the effectiveness of our learned descriptors for automatic texturing, material-aware retrieval, and physical simulation. The dataset and code will be publicly available.
30.Synscapes: A Photorealistic Synthetic Dataset for Street Scene Parsing pdf
We introduce Synscapes -- a synthetic dataset for street scene parsing created using photorealistic rendering techniques, and show state-of-the-art results for training and validation as well as new types of analysis. We study the behavior of networks trained on real data when performing inference on synthetic data: a key factor in determining the equivalence of simulation environments. We also compare the behavior of networks trained on synthetic data and evaluated on real-world data. Additionally, by analyzing pre-trained, existing segmentation and detection models, we illustrate how uncorrelated images along with a detailed set of annotations open up new avenues for analysis of computer vision systems, providing fine-grain information about how a model's performance changes according to factors such as distance, occlusion and relative object orientation.
31.Understanding Deep Convolutional Networks through Gestalt Theory pdf
The superior performance of deep convolutional networks over high-dimensional problems have made them very popular for several applications. Despite their wide adoption, their underlying mechanisms still remain unclear with their improvement procedures still relying mainly on a trial and error process. We introduce a novel sensitivity analysis based on the Gestalt theory for giving insights into the classifier function and intermediate layers. Since Gestalt psychology stipulates that perception can be a product of complex interactions among several elements, we perform an ablation study based on this concept to discover which principles and image context significantly contribute in the network classification. Our results reveal that ConvNets follow most of the visual cortical perceptual mechanisms defined by the Gestalt principles at several levels. The proposed framework stimulates specific feature maps in classification problems and reveal important network attributes that can produce more explainable network models.
32.Fast and Robust Multiple ColorChecker Detection using Deep Convolutional Neural Networks pdf
ColorCheckers are reference standards that professional photographers and filmmakers use to ensure predictable results under every lighting condition. The objective of this work is to propose a new fast and robust method for automatic ColorChecker detection. The process is divided into two steps: (1) ColorCheckers localization and (2) ColorChecker patches recognition. For the ColorChecker localization, we trained a detection convolutional neural network using synthetic images. The synthetic images are created with the 3D models of the ColorChecker and different background images. The output of the neural networks are the bounding box of each possible ColorChecker candidates in the input image. Each bounding box defines a cropped image which is evaluated by a recognition system, and each image is canonized with regards to color and dimensions. Subsequently, all possible color patches are extracted and grouped with respect to the center's distance. Each group is evaluated as a candidate for a ColorChecker part, and its position in the scene is estimated. Finally, a cost function is applied to evaluate the accuracy of the estimation. The method is tested using real and synthetic images. The proposed method is fast, robust to overlaps and invariant to affine projections. The algorithm also performs well in case of multiple ColorCheckers detection.
33.Can We Gain More from Orthogonality Regularizations in Training Deep CNNs? pdf
This paper seeks to answer the question: as the (near-) orthogonality of weights is found to be a favorable property for training deep convolutional neural networks, how can we enforce it in more effective and easy-to-use ways? We develop novel orthogonality regularizations on training deep CNNs, utilizing various advanced analytical tools such as mutual coherence and restricted isometry property. These plug-and-play regularizations can be conveniently incorporated into training almost any CNN without extra hassle. We then benchmark their effects on state-of-the-art models: ResNet, WideResNet, and ResNeXt, on several most popular computer vision datasets: CIFAR-10, CIFAR-100, SVHN and ImageNet. We observe consistent performance gains after applying those proposed regularizations, in terms of both the final accuracies achieved, and faster and more stable convergences. We have made our codes and pre-trained models publicly available: this https URL.