References & Citations
Computer Science > Computer Vision and Pattern Recognition
Title: CrossRectify: Leveraging Disagreement for Semi-supervised Object Detection
(Submitted on 26 Jan 2022 (v1), last revised 18 Apr 2022 (this version, v2))
Abstract: Semi-supervised object detection has recently achieved substantial progress. As a mainstream solution, the self-labeling-based methods train the detector on both labeled data and unlabeled data with pseudo labels predicted by the detector itself, but their performances are always limited. Through experimental analysis, we reveal the underlying reason is that the detector is misguided by the incorrect pseudo labels predicted by itself (dubbed self-errors). These self-errors can hurt performance even worse than random-errors, and can be neither discerned nor rectified during the self-labeling process. In this paper, we propose an effective detection framework named CrossRectify, to obtain accurate pseudo labels by simultaneously training two detectors with different initial parameters. Specifically, the proposed approach leverages the disagreements between detectors to discern the self-errors and refines the pseudo label quality by the proposed cross-rectifying mechanism. Extensive experiments show that CrossRectify achieves outperforming performances over various detector structures on 2D and 3D detection benchmarks.
Submission history
From: Chengcheng Ma [view email][v1] Wed, 26 Jan 2022 03:34:57 GMT (3387kb,D)
[v2] Mon, 18 Apr 2022 03:58:23 GMT (1773kb,D)
Link back to: arXiv, form interface, contact.