Knowledge Commons of Institute of Automation,CAS
Large Batch Optimization for Object Detection: Training COCO in 12 minutes | |
Wang, Tong1,2![]() ![]() ![]() ![]() ![]() | |
2020-08 | |
会议名称 | European Conference on Compter Vision (ECCV) |
页码 | 481-496 |
会议日期 | 2020-8-24 |
会议地点 | Online |
摘要 | Most of existing object detectors usually adopt a small training batch size (e.g. 16), which severely hinders the whole community from exploring large-scale datasets due to the extremely long training procedure. In this paper, we propose a versatile large batch optimization framework for object detection, named LargeDet, which successfully scales the batch size to larger than 1K for the first time. Specifically, we present a novel Periodical Moments Decay LAMB (PMD-LAMB) algorithm to effectively reduce the negative effects of the lagging historical gradients. Additionally, the Synchronized Batch Normalization (SyncBN) is utilized to help fast convergence. With LargeDet, we can not only prominently shorten the training period, but also significantly improve the detection accuracy of sparsely annotated large-scale datasets. For instance, we can finish the training of ResNet50 FPN detector on COCO within 12 minutes. Moreover, we achieve 12.2% mAP@0.5 absolute improvement for ResNet50 FPN on Open Images by training with batch size 640. |
关键词 | Object detection Large batch optimization Periodical moments decay |
学科领域 | 模式识别 |
学科门类 | 工学::计算机科学与技术(可授工学、理学学位) |
收录类别 | EI |
资助项目 | National Natural Science Foundation of China (NSFC)[61633002] ; National Nature Science Foundation of China[61876086] ; National Natural Science Foundation of China[61702510] ; National Natural Science Foundation of China[61806200] ; National Natural Science Foundation of China[61772527] ; National Natural Science Foundation of China[61772527] ; National Natural Science Foundation of China[61806200] ; National Natural Science Foundation of China[61702510] ; National Nature Science Foundation of China[61876086] ; National Natural Science Foundation of China (NSFC)[61633002] |
语种 | 英语 |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/47416 |
专题 | 紫东太初大模型研究中心_图像与视频分析 |
作者单位 | 1.National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, Beijing, China 2.School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing, China 3.ObjectEye Inc., Beijing, China 4.Peking University, Beijing, China 5.Peng Cheng Laboratory, Shenzhen, China 6.NEXWISE Co., Ltd, Guangzhou, China |
第一作者单位 | 模式识别国家重点实验室 |
推荐引用方式 GB/T 7714 | Wang, Tong,Zhu, Yousong,Zhao, Chaoyang,et al. Large Batch Optimization for Object Detection: Training COCO in 12 minutes[C],2020:481-496. |
条目包含的文件 | 下载所有文件 | |||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
Wang2020_Chapter_Lar(3706KB) | 会议论文 | 开放获取 | CC BY-NC-SA | 浏览 下载 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论