Focal and efficient IOU loss for accurate bounding box regression | |
Zhang, Yi-Fan1,2,3; Ren, Weiqiang4; Zhang, Zhang1,2,3; Jia, Zhen1,2; Wang, Liang1,2,3; Tan, Tieniu1,2,3 | |
发表期刊 | NEUROCOMPUTING |
ISSN | 0925-2312 |
2022-09-28 | |
卷号 | 506页码:146-157 |
通讯作者 | Zhang, Zhang(zzhang@nlpr.ia.ac.cn) |
摘要 | In object detection, bounding box regression (BBR) is a crucial step that determines the object localization performance. However, we find that most previous loss functions for BBR have two main drawbacks: (i) Both `n-norm and IOU-based loss functions are inefficient to depict the objective of BBR, which leads to slow convergence and inaccurate regression results. (ii) Most of the loss functions ignore the imbalance problem in BBR that the large number of anchor boxes which have small overlaps with the target boxes contribute most to the optimization of BBR. To mitigate the adverse effects caused thereby, we perform thorough studies to exploit the potential of BBR losses in this paper. Firstly, an Efficient Intersection over Union (EIOU) loss is proposed, which explicitly measures the discrepancies of three geometric factors in BBR, i.e., the overlap area, the central point and the side length. After that, we state the Effective Example Mining (EEM) problem and propose a regression version of focal loss to make the regression process focus on high-quality anchor boxes. Finally, the above two parts are combined to obtain a new loss function, namely Focal-EIOU loss. Extensive experiments on both synthetic and real datasets are performed. Notable superiorities on both the convergence speed and the localization accuracy can be achieved over other BBR losses. (c) 2022 Elsevier B.V. All rights reserved. |
关键词 | Object detection Loss function design Hard sample mining |
DOI | 10.1016/j.neucom.2022.07.042 |
收录类别 | SCI |
语种 | 英语 |
WOS研究方向 | Computer Science |
WOS类目 | Computer Science, Artificial Intelligence |
WOS记录号 | WOS:000914170300012 |
出版者 | ELSEVIER |
引用统计 | |
文献类型 | 期刊论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/51384 |
专题 | 智能感知与计算研究中心 |
通讯作者 | Zhang, Zhang |
作者单位 | 1.Chinese Acad Sci CASIA, Inst Automat, CRIPAC, Beijing, Peoples R China 2.Chinese Acad Sci CASIA, Inst Automat, NLPR, Beijing, Peoples R China 3.Univ Chinese Acad Sci, Beijing, Peoples R China 4.Horizon Robot, Beijing, Peoples R China |
第一作者单位 | 中国科学院自动化研究所; 模式识别国家重点实验室 |
通讯作者单位 | 中国科学院自动化研究所; 模式识别国家重点实验室 |
推荐引用方式 GB/T 7714 | Zhang, Yi-Fan,Ren, Weiqiang,Zhang, Zhang,et al. Focal and efficient IOU loss for accurate bounding box regression[J]. NEUROCOMPUTING,2022,506:146-157. |
APA | Zhang, Yi-Fan,Ren, Weiqiang,Zhang, Zhang,Jia, Zhen,Wang, Liang,&Tan, Tieniu.(2022).Focal and efficient IOU loss for accurate bounding box regression.NEUROCOMPUTING,506,146-157. |
MLA | Zhang, Yi-Fan,et al."Focal and efficient IOU loss for accurate bounding box regression".NEUROCOMPUTING 506(2022):146-157. |
条目包含的文件 | 条目无相关文件。 |
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