Boosting Weakly Supervised Object Detection via Learning Bounding Box Adjusters

Weakly-supervised object detection (WSOD) has emerged as an inspiring recent topic to avoid expensive instance-level object annotations. However, the bounding boxes of most existing WSOD methods are mainly determined by precomputed proposals, thereby being limited in precise object localization. In...

Full description

Saved in:
Bibliographic Details
Published inProceedings / IEEE International Conference on Computer Vision pp. 2856 - 2865
Main Authors Dong, Bowen, Huang, Zitong, Guo, Yuelin, Wang, Qilong, Niu, Zhenxing, Zuo, Wangmeng
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.10.2021
Subjects
Classification algorithms
Codes
Computational modeling
Computer vision
Detection and localization in 2D and 3D
Location awareness
Object detection
Training
Online AccessGet full text
ISSN2380-7504
DOI10.1109/ICCV48922.2021.00287

Cover

More Information
Summary:Weakly-supervised object detection (WSOD) has emerged as an inspiring recent topic to avoid expensive instance-level object annotations. However, the bounding boxes of most existing WSOD methods are mainly determined by precomputed proposals, thereby being limited in precise object localization. In this paper, we defend the problem setting for improving localization performance by leveraging the bounding box regression knowledge from a well-annotated auxiliary dataset. First, we use the well-annotated auxiliary dataset to explore a series of learnable bounding box adjusters (LBBAs) in a multi-stage training manner, which is class-agnostic. Then, only LBBAs and a weakly-annotated dataset with non-overlapped classes are used for training LBBA-boosted WSOD. As such, our LBBAs are practically more convenient and economical to implement while avoiding the leakage of the auxiliary well-annotated dataset. In particular, we formulate learning bounding box adjusters as a bi-level optimization problem and suggest an EM-like multi-stage training algorithm. Then, a multi-stage scheme is further presented for LBBA-boosted WSOD. Additionally, a masking strategy is adopted to improve proposal classification. Experimental results verify the effectiveness of our method. Our method performs favorably against state-of-the-art WSOD methods and knowledge transfer model with similar problem setting. Code is publicly available at https://github.com/DongSky/lbba_boosted_wsod.
ISSN:2380-7504
DOI:10.1109/ICCV48922.2021.00287