CrossRectify: Leveraging disagreement for semi-supervised object detection

•We point out that the performances of self-labeling-based semi-supervised object detection (SSOD) approaches are always limited, and the reason behind such phenomenon lies in that the detector can neither discern nor rectify the misclassified pseudo bounding boxes predicted by itself.•We propose an effective approach named CrossRectify to discern and rectify the misclassified pseudo bounding boxes using the disagreements between two detectors, which can address the inherent limitations of self-labeling and improve the detection performance.•We conduct extensive experiments on both 2D and 3D object detection benchmark datasets, and the results verify the superiority of the proposed CrossRectify approach, compared with state-of-the-art approaches. 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.