arrow
返回

Discrepant multiple instance learning for weakly supervised object detection

delete2022-02-01
delete30
PRE
AI
高伟 封面图
高伟 (Wei Gao)
F
Fang Wan *
J
Jun Yue
S
Songcen Xu
Q
Qixiang Ye
DOI:10.1016/j.patcog.2021.108233delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
Multiple Instance Learning (MIL) is a fundamental method for weakly supervised object detection (WSOD), but experiences difficulty in excluding local optimal solutions and may miss objects or falsely localize object parts. In this paper, we introduce discrepantly collaborative modules into MIL and thereby create discrepant multiple instance learning (D-MIL), pursuing optimal solutions in a simple-yet-effective way. D-MIL adopts multiple MIL learners to pursue discrepant yet complementary solutions indicating object parts, which are fused with a collaboration module for precise object localization. D-MIL implements a new teachers-students model, where MIL learners act as teachers and object detectors as students. Multiple teachers provide rich yet complementary information, which are absorbed by students and transferred back to reinforce the performance of teachers. Experiments show that D-MIL significantly improves the baseline while achieves state-of-the-art performance on the challenging MS-COCO object detection benchmark. (c) 2021 Elsevier Ltd. All rights reserved.
Keyword:
Weakly supervised detection
Multiple instance learning
Learner discrepancy
Collaborative learning

期刊

Pattern Recognition 封面图
Pattern Recognition
IF:
7.6
论文数:
1.3W
被引数:
4.5W

机构

U
university of chinese academy of sciences, cas
学者数:
4.1W
论文数: 3.8W
被引数: 75
C
chinese academy of sciences
学者数:
56.7W
论文数: 44.9W
被引数: 704