Res2Net for object detection and instance segmentation based on mmdetection.

News

• Res2Net is now one of the basic bonebones in MMDetection v2 framework https://github.com/open-mmlab/mmdetection. Using MMDetection v2 with Res2Net achieves better performance with less computational cost.
• Res2Net with mmdetection v2 is now available in https://github.com/Res2Net/mmdetection/tree/mmdet_res2net.

Introduction

We propose a novel building block for CNNs, namely Res2Net, by constructing hierarchical residual-like connections within one single residual block. The Res2Net represents multi-scale features at a granular level and increases the range of receptive fields for each network layer.

Backbone	Params.	GFLOPs	top-1 err.	top-5 err.
ResNet-101	44.6 M	7.8	22.63	6.44
ResNeXt-101-64x4d	83.5M	15.5	20.40	-
HRNetV2p-W48	77.5M	16.1	20.70	5.50
Res2Net-101	45.2M	8.3	18.77	4.64

Compared with other backbone networks, Res2Net requires fewer parameters and FLOPs.

Note:

• GFLOPs for classification are calculated with image size (224x224).

Detection and segmentation Results

Faster R-CNN

Backbone	Params.	GFLOPs	box AP
R-101-FPN	60.52M	283.14	39.4
X-101-64x4d-FPN	99.25M	440.36	41.3
HRNetV2p-W48	83.36M	459.66	41.5
Res2Net-101	61.18M	293.68	42.3

Mask R-CNN

Backbone	Params.	GFLOPs	box AP	mask AP
R-101-FPN	63.17M	351.65	40.3	36.5
X-101-64x4d-FPN	101.9M	508.87	42.0	37.7
HRNetV2p-W48	86.01M	528.17	42.9	38.3
Res2Net-101	63.83M	362.18	43.3	38.6

Cascade R-CNN

Backbone	Params.	GFLOPs	box AP
R-101-FPN	88.16M	310.78	42.5
X-101-64x4d-FPN	126.89M	468.00	44.7
HRNetV2p-W48	111.00M	487.30	44.6
Res2Net-101	88.82M	321.32	45.5

Cascade Mask R-CNN

Backbone	Params.	GFLOPs	box AP	mask AP
R-101-FPN	96.09M	516.30	43.3	37.6
X-101-64x4d-FPN	134.82M	673.52	45.7	39.4
HRNetV2p-W48	118.93M	692.82	46.0	39.5
Res2Net-101	96.75M	526.84	46.1	39.4

Hybrid Task Cascade (HTC)

Backbone	Params.	GFLOPs	box AP	mask AP
R-101-FPN	99.03M	563.76	44.9	39.4
X-101-64x4d-FPN	137.75M	720.98	46.9	40.8
HRNetV2p-W48	121.87M	740.28	47.0	41.0
Res2Net-101	99.69M	574.30	47.5	41.3

Note:

• GFLOPs are calculated with image size (1280, 800).
• All detection methods in this page use pytorch style. Lr schd is 2x for Faster R-CNN and Mask R-CNN, and 20e for others.
• Res2Net ImageNet pretrained models are in Res2Net-PretrainedModels.
• More applications of Res2Net are in Res2Net-Github.

MMDetection

News: We released the technical report on ArXiv.

Documentation: https://mmdetection.readthedocs.io/

Introduction

The master branch works with PyTorch 1.1 to 1.4.

mmdetection is an open source object detection toolbox based on PyTorch. It is a part of the open-mmlab project developed by Multimedia Laboratory, CUHK.

Major features

• Modular Design

  We decompose the detection framework into different components and one can easily construct a customized object detection framework by combining different modules.

• Support of multiple frameworks out of box

  The toolbox directly supports popular and contemporary detection frameworks, e.g. Faster RCNN, Mask RCNN, RetinaNet, etc.

• High efficiency

  All basic bbox and mask operations run on GPUs now. The training speed is faster than or comparable to other codebases, including Detectron, maskrcnn-benchmark and SimpleDet.

• State of the art

  The toolbox stems from the codebase developed by the MMDet team, who won COCO Detection Challenge in 2018, and we keep pushing it forward.

Apart from MMDetection, we also released a library mmcv for computer vision research, which is heavily depended on by this toolbox.

License

This project is released under the Apache 2.0 license.

Changelog

v1.1.0 was released in 24/2/2020. Please refer to CHANGELOG.md for details and release history.

Benchmark and model zoo

Supported methods and backbones are shown in the below table. Results and models are available in the Model zoo.

	ResNet	ResNeXt	SENet	VGG	HRNet	Res2Net
RPN	✓	✓	☐	✗	✓	✓
Fast R-CNN	✓	✓	☐	✗	✓	✓
Faster R-CNN	✓	✓	☐	✗	✓	✓
Mask R-CNN	✓	✓	☐	✗	✓	✓
Cascade R-CNN	✓	✓	☐	✗	✓	✓
Cascade Mask R-CNN	✓	✓	☐	✗	✓	✓
SSD	✗	✗	✗	✓	✗	✗
RetinaNet	✓	✓	☐	✗	✓	✓
GHM	✓	✓	☐	✗	✓	✓
Mask Scoring R-CNN	✓	✓	☐	✗	✓	✓
Double-Head R-CNN	✓	✓	☐	✗	✓	✓
Grid R-CNN (Plus)	✓	✓	☐	✗	✓	✓
Hybrid Task Cascade	✓	✓	☐	✗	✓	✓
Libra R-CNN	✓	✓	☐	✗	✓	✓
Guided Anchoring	✓	✓	☐	✗	✓	✓
FCOS	✓	✓	☐	✗	✓	✓
RepPoints	✓	✓	☐	✗	✓	✓
Foveabox	✓	✓	☐	✗	✓	✓
FreeAnchor	✓	✓	☐	✗	✓	✓
NAS-FPN	✓	✓	☐	✗	✓	✓
ATSS	✓	✓	☐	✗	✓	✓

Other features

• CARAFE
• DCNv2
• Group Normalization
• Weight Standardization
• OHEM
• Soft-NMS
• Generalized Attention
• GCNet
• Mixed Precision (FP16) Training
• InstaBoost

Installation

Please refer to INSTALL.md for installation and dataset preparation.

Get Started

Please see GETTING_STARTED.md for the basic usage of MMDetection.

Contributing

We appreciate all contributions to improve MMDetection. Please refer to CONTRIBUTING.md for the contributing guideline.

Acknowledgement

MMDetection is an open source project that is contributed by researchers and engineers from various colleges and companies. We appreciate all the contributors who implement their methods or add new features, as well as users who give valuable feedbacks. We wish that the toolbox and benchmark could serve the growing research community by providing a flexible toolkit to reimplement existing methods and develop their own new detectors.

Citation

If you use this toolbox or benchmark in your research, please cite this project.

@article{mmdetection,
  title   = {{MMDetection}: Open MMLab Detection Toolbox and Benchmark},
  author  = {Chen, Kai and Wang, Jiaqi and Pang, Jiangmiao and Cao, Yuhang and
             Xiong, Yu and Li, Xiaoxiao and Sun, Shuyang and Feng, Wansen and
             Liu, Ziwei and Xu, Jiarui and Zhang, Zheng and Cheng, Dazhi and
             Zhu, Chenchen and Cheng, Tianheng and Zhao, Qijie and Li, Buyu and
             Lu, Xin and Zhu, Rui and Wu, Yue and Dai, Jifeng and Wang, Jingdong
             and Shi, Jianping and Ouyang, Wanli and Loy, Chen Change and Lin, Dahua},
  journal= {arXiv preprint arXiv:1906.07155},
  year={2019}
}

Contact

This repo is currently maintained by Kai Chen (@hellock), Yuhang Cao (@yhcao6), Wenwei Zhang (@ZwwWayne), Jiarui Xu (@xvjiarui). Other core developers include Jiangmiao Pang (@OceanPang) and Jiaqi Wang (@myownskyW7).