MMDetection: Open MMLab Detection Toolbox and Benchmark
Kai Chen1 Jiaqi Wang1 Jiangmiao Pang2∗ Yuhang Cao1 Yu Xiong1 Xiaoxiao Li1 Shuyang Sun3 Wansen Feng4 Ziwei Liu1 Jiarui Xu5 Zheng Zhang6 Dazhi Cheng7 Chenchen Zhu8 Tianheng Cheng9 Qijie Zhao10 Buyu Li1 Xin Lu4 Rui Zhu11 Yue Wu12 Jifeng Dai6 Jingdong Wang6 Jianping Shi4 Wanli Ouyang3 Chen Change Loy13 Dahua Lin1 1The Chinese University of Hong Kong 2Zhejiang University 3The University of Sydney 4SenseTime Research 5Hong Kong University of Science and Technology 6Microsoft Research Asia 7Beijing Institute of Technology 8Nanjing University 9Huazhong University of Science and Technology 10Peking University 11Sun Yat-sen University 12Northeastern University 13Nanyang Technological University indicates equal contribution.
Abstract
We present MMDetection, an object detection toolbox that contains a rich set of object detection and instance segmentation methods as well as related components and modules. The toolbox started from a codebase of MMDet team who won the detection track of COCO Challenge 2018. It gradually evolves into a unified platform that covers many popular detection methods and contemporary modules. It not only includes training and inference codes, but also provides weights for more than 200 network models. We believe this toolbox is by far the most complete detection toolbox. In this paper, we introduce the various features of this toolbox. In addition, we also conduct a benchmarking study on different methods, components, and their hyper-parameters. 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. Code and models are available at https://github.com/open-mmlab/mmdetection. The project is under active development and we will keep this document updated.
中文速览
目标检测(object detection)和实例分割(instance segmentation)是计算机视觉的核心任务，但不同代码实现之间的差异往往让结果难以复现和公平比较。MMDetection 正是为此而生——它以模块化设计将检测框架拆解为骨干网络、颈部、检测头等独立组件，让研究者可以像搭积木一样灵活组合，一套代码就能支持从 SSD、RetinaNet 到 Cascade R-CNN、Mask R-CNN 在内的数十种主流方法，并提供超过 200 个预训练模型权重。实验表明，MMDetection 在速度、显存占用和检测精度上均优于或持平于 Detectron、maskrcnn-benchmark 等同类工具箱，且支持混合精度训练和多节点近线性加速。此外，论文还系统比较了多种回归损失函数、归一化层和训练策略的实际效果，为社区提供了可复现的基准和最佳实践参考，有助于推动目标检测研究走向更规范、更高效的方向。
原文 arXiv:1906.07155;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1906.07155v1