Source-Free Unsupervised Cross-Domain Pedestrian Detection via Pseudo Label Mining and Screening

Zhiri Tang, Qianfen Jiao, Jian Zhong, Si Wu, Hau-San WongICME 2022 · OpenReview

Abstract

Although current cross-domain pedestrian detection frame-works have obtained certain positive results, the performance is still source data dependent, which is cumbersome and im-practical in practical applications. To address this issue, we propose a source-free unsupervised pedestrian detection with pseudo label mining and screening. First, a modified CSP de-tector with DropBlock and three detection heads is presented. Then, a multi-expert method is proposed to fuse pseudo la-bels from three detection heads. Finally, a clustering-based self-supervised learning is adopted to categorize pseudo la-bels into positive and negative classes, which forms a set of clusters via similarity of pseudo labels and give classification results based on two confidence scores of each label from the detector backbone and multi-expert fusion. Experimental re-sults on three benchmark datasets show that the proposed approach can achieve state-of-the-art performance and be even comparable with other latest works using source data.
BibTeX
@inproceedings{tang2022source,
  author = {Zhiri Tang and Qianfen Jiao and Jian Zhong and Si Wu and Hau-San Wong},
  title = {{Source-Free Unsupervised Cross-Domain Pedestrian Detection via Pseudo Label Mining and Screening}},
  booktitle = {ICME},
  year = {2022},
  url = {https://openreview.net/forum?id=MbGKudBHPd},
}

Loading related content...

Figure