English

Predicting Class Distribution Shift for Reliable Domain Adaptive Object Detection

Computer Vision and Pattern Recognition 2023-08-29 v2 Robotics

Abstract

Unsupervised Domain Adaptive Object Detection (UDA-OD) uses unlabelled data to improve the reliability of robotic vision systems in open-world environments. Previous approaches to UDA-OD based on self-training have been effective in overcoming changes in the general appearance of images. However, shifts in a robot's deployment environment can also impact the likelihood that different objects will occur, termed class distribution shift. Motivated by this, we propose a framework for explicitly addressing class distribution shift to improve pseudo-label reliability in self-training. Our approach uses the domain invariance and contextual understanding of a pre-trained joint vision and language model to predict the class distribution of unlabelled data. By aligning the class distribution of pseudo-labels with this prediction, we provide weak supervision of pseudo-label accuracy. To further account for low quality pseudo-labels early in self-training, we propose an approach to dynamically adjust the number of pseudo-labels per image based on model confidence. Our method outperforms state-of-the-art approaches on several benchmarks, including a 4.7 mAP improvement when facing challenging class distribution shift.

Keywords

Cite

@article{arxiv.2302.06039,
  title  = {Predicting Class Distribution Shift for Reliable Domain Adaptive Object Detection},
  author = {Nicolas Harvey Chapman and Feras Dayoub and Will Browne and Christopher Lehnert},
  journal= {arXiv preprint arXiv:2302.06039},
  year   = {2023}
}
R2 v1 2026-06-28T08:38:16.363Z