In this paper, we address the limitations of the DETR-based semi-supervised object detection (SSOD) framework (Zhang et al, 2023), particularly focusing on the challenges posed by the quality of object queries. In DETR-based SSOD, the one-to-one assignment strategy provides inaccurate pseudo-labels, while the one-to-many assignments strategy leads to overlapping predictions. These issues compromise training efficiency and degrade model performance, especially in detecting small or occluded objects. We introduce Sparse Semi-DETR, a novel transformer-based, end-to-end semi-supervised object detection solution to overcome these challenges. Sparse Semi-DETR incorporates a Query Refinement Module to enhance the quality of object queries, significantly improving detection capabilities for small and partially obscured objects.
Additionally, we integrate a Reliable Pseudo-Label Filtering Module that selectively filters high-quality pseudo-labels, thereby enhancing detection accuracy and consistency. On the MS-COCO and Pascal VOC object detection benchmarks, Sparse Semi-DETR achieves a significant improvement over current state-of-the-art methods that highlight Sparse Semi-DETR's effectiveness in semi-supervised object detection, particularly in challenging scenarios involving small or partially obscured objects.

paper link: https://openaccess.thecvf.com/content/CVPR2024/papers/Shehzadi_Sparse_Semi-DETR_Sparse_Learnable_Queries_for_Semi-Supervised_Object_Detection_CVPR_2024_paper.pdf
slide: https://cvpr.thecvf.com/media/cvpr-2024/Slides/30138.pdf