• Complex
  • Title
  • Keyword
  • Abstract
  • Scholars
  • Journal
  • ISSN
  • Conference
成果搜索

author:

Cheng, Qi (Cheng, Qi.) [1] | Wu, Yingjie (Wu, Yingjie.) [2] | Chen, Fei (Chen, Fei.) [3] (Scholars:陈飞) | Guo, Yilong (Guo, Yilong.) [4]

Indexed by:

EI Scopus

Abstract:

In object detection, the imbalance problem often occurs when the number of training samples of different categories varies greatly, or multiple loss functions need to be minimized which is harmful to the performance of the detector. In this paper, we consider that the imbalance problem can be implied by the imbalance of gradient distribution. To address these imbalance issues, we analyze the gradient of cross-entropy loss and propose balanced cross-entropy (BLCE) loss and balanced binary cross-entropy (BBCE) loss for solving objective imbalance and class imbalance issues respectively. The BLCE loss significantly reduces the overall classification loss and keeps the classification loss and regression loss balanced. Furthermore, the BBCE loss automatically down-weight the contribution of inliers during training and rapidly focus the model on outliers. Ablation studies on object detection and image classification demonstrate the effectiveness of our loss function. We replace the corresponding losses in Libra R-CNN and evaluate our detector on the COCO test-dev. Our results show that Libra R-CNN can surpass the accuracy of many existing state-of-the-art detectors when training with our balanced loss. © 2020, Springer Nature Switzerland AG.

Keyword:

Computer vision Entropy Object detection Object recognition

Community:

  • [ 1 ] [Cheng, Qi]College of Mathematics and Computer Science, Fuzhou University, Fuzhou; 350108, China
  • [ 2 ] [Wu, Yingjie]College of Mathematics and Computer Science, Fuzhou University, Fuzhou; 350108, China
  • [ 3 ] [Chen, Fei]College of Mathematics and Computer Science, Fuzhou University, Fuzhou; 350108, China
  • [ 4 ] [Guo, Yilong]College of Mathematics and Computer Science, Fuzhou University, Fuzhou; 350108, China

Reprint 's Address:

  • 吴英杰

    [wu, yingjie]college of mathematics and computer science, fuzhou university, fuzhou; 350108, china

Show more details

Version:

Related Keywords:

Related Article:

Source :

ISSN: 0302-9743

Year: 2020

Volume: 12307 LNCS

Page: 342-354

Language: English

0 . 4 0 2

JCR@2005

Cited Count:

WoS CC Cited Count: 0

SCOPUS Cited Count:

ESI Highly Cited Papers on the List: 0 Unfold All

WanFang Cited Count:

Chinese Cited Count:

30 Days PV: 2

Online/Total:151/10011164
Address:FZU Library(No.2 Xuyuan Road, Fuzhou, Fujian, PRC Post Code:350116) Contact Us:0591-22865326
Copyright:FZU Library Technical Support:Beijing Aegean Software Co., Ltd. 闽ICP备05005463号-1