Distribution-Aware Calibration for Object Detection with Noisy Bounding Boxes


Donghao Zhou (The Chinese University of Hong Kong), Jialin Li (Tencent YouTu Lab), Jinpeng Li (The Chinese University of Hong Kong), Jiancheng Huang (Chinese Academy of Sciences), Qiang Nie (The Hong Kong University of Science and Technology), Yong Liu (Tencent Youtu Lab), Bin-Bin Gao (Tencent), Qiong Wang (Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Chinese Academy of Sciences), Pheng-Ann Heng (The Chinese University of Hong Kong), Guangyong Chen (Zhejiang Lab)
The 35th British Machine Vision Conference
PDFPosterSupplementary

Abstract

Large-scale well-annotated datasets are of great importance for training an effective object detector. However, obtaining accurate bounding box annotations is laborious and demanding. Unfortunately, the resultant noisy bounding boxes could cause corrupt supervision signals and thus diminish detection performance. Motivated by the observation that the real ground-truth is usually situated in the aggregation region of the proposals assigned to a noisy ground-truth, we propose DIStribution-aware CalibratiOn (DISCO) to model the spatial distribution of proposals for calibrating supervision signals. In DISCO, spatial distribution modeling is performed to statistically extract the potential locations of objects. Based on the modeled distribution, three distribution-aware techniques, i.e., distribution-aware proposal augmentation (DA-Aug), distribution-aware box refinement (DA-Ref), and distribution-aware confidence estimation (DA-Est), are developed to improve classification, localization, and interpretability, respectively. Extensive experiments demonstrate that DISCO can achieve SOTA performance in this task, especially at high noise levels. Code is available at https://github.com/Correr-Zhou/DISCO.

Citation

@inproceedings{Zhou_2024_BMVC,
author    = {Donghao Zhou and Jialin Li and Jinpeng Li and Jiancheng Huang and Qiang Nie and Yong Liu and Bin-Bin Gao and Qiong Wang and Pheng-Ann Heng and Guangyong Chen},
title     = {Distribution-Aware Calibration for Object Detection with Noisy Bounding Boxes},
booktitle = {35th British Machine Vision Conference 2024, {BMVC} 2024, Glasgow, UK, November 25-28, 2024},
publisher = {BMVA},
year      = {2024},
url       = {https://papers.bmvc2024.org/0102.pdf}
}


Copyright © 2024 The British Machine Vision Association and Society for Pattern Recognition
The British Machine Vision Conference is organised by The British Machine Vision Association and Society for Pattern Recognition. The Association is a Company limited by guarantee, No.2543446, and a non-profit-making body, registered in England and Wales as Charity No.1002307 (Registered Office: Dept. of Computer Science, Durham University, South Road, Durham, DH1 3LE, UK).

Imprint | Data Protection