Institutional Repository of Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
C2AM Loss: Chasing a Better Decision Boundary for Long-Tail Object Detection | |
Wang, Tong1,2![]() ![]() ![]() ![]() ![]() ![]() ![]() | |
2022-06 | |
Conference Name | IEEE Computer Vision and Pattern Recognition (CVPR) |
Conference Date | 2022-6-19 |
Conference Place | New Orleans, Louisiana & Online |
Abstract | Long-tail object detection suffers from poor performance on tail categories. We reveal that the real culprit lies in the extremely imbalanced distribution of the classifier’s weight norm. For conventional softmax cross-entropy loss, such imbalanced weight norm distribution yields ill-conditioned decision boundary for categories which have small weight |
Subject Area | 模式识别 |
MOST Discipline Catalogue | 工学::计算机科学与技术(可授工学、理学学位) |
Indexed By | EI |
Funding Project | National Nature Science Foundation of China[61876086] ; National Nature Science Foundation of China[61876086] |
Language | 英语 |
Sub direction classification | 目标检测、跟踪与识别 |
Document Type | 会议论文 |
Identifier | http://ir.ia.ac.cn/handle/173211/47418 |
Collection | 模式识别国家重点实验室_图像与视频分析 |
Affiliation | 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.Peng Cheng Laboratory, Shenzhen, China 4.Development Research Institute of Guangzhou Smart City, Guangzhou, China |
First Author Affilication | Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China |
Recommended Citation GB/T 7714 | Wang, Tong,Zhu, Yousong,Chen, Yingying,et al. C2AM Loss: Chasing a Better Decision Boundary for Long-Tail Object Detection[C],2022. |
Files in This Item: | Download All | |||||
File Name/Size | DocType | Version | Access | License | ||
C2AM Loss Chasing a (5757KB) | 会议论文 | 开放获取 | CC BY-NC-SA | View Download |
Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.
Edit Comment