Tail Classes Matter: Long-Tailed Object Detection Revisited
Yinglu Zhang, Chenbo Zhang, Lu Zhang, Tianying Liu, Jihong Guan, Xinkai Liang, Jiajia Zhao, Shuigeng Zhou
Abstract
Real-world data ubiquitously exhibit long-tailed distribution, which sparks the increasing interest in long-tailed object detection (LTOD). However, existing methods neglect that a lack of diverse data in tail classes will cause underrepresented tail class features, making their efforts for balancing foreground classes tend to over-fit tail classes and be less effective. In this paper, we propose a multi-class co-attention generation network to increase data diversity of tail classes by generating augmented samples. To alleviate imbalance, we develop a distribution-aware up-sampling strategy, performing differential up-sampling for different classes and design a bi-directional regulation loss to adjust both positive and negative gradients. Moreover, we construct a new dataset LVIS-X with more rare classes based on existing LTOD benchmark dataset LVIS. Experiments on LVIS and LVIS-X demonstrate the superiority of the proposed method.
BibTeX
@inproceedings{icassp2024_tailclassesmatte,
title = {Tail Classes Matter: Long-Tailed Object Detection Revisited},
author = {Yinglu Zhang and Chenbo Zhang and Lu Zhang and Tianying Liu and Jihong Guan and Xinkai Liang and Jiajia Zhao and Shuigeng Zhou},
booktitle = {ICASSP 2024},
year = {2024}
}