论文标题
S4OD：单阶段对象检测的半监督学习
S4OD: Semi-Supervised learning for Single-Stage Object Detection
论文作者
论文摘要
单阶段探测器患有极端的前景阶级失衡，而两阶段探测器则没有。因此，在半监督对象检测中，两阶段检测器只能通过基于分类分数选择高质量的伪标签来提供出色的性能。但是，将此策略直接应用于单阶段探测器将加剧类不平衡的阳性样本。因此，单阶段探测器必须同时考虑伪标签的质量和数量。在本文中，我们在分类部门设计了动态的自适应阈值（DSAT）策略，该策略可以自动选择伪标签以实现质量和数量之间的最佳权衡。此外，为了评估单阶段探测器中伪标签的回归质量，我们提出了一个模块来计算基于非最大抑制作用的盒子的回归不确定性。通过仅利用10％的可可标记数据，我们的方法在无锚定检测器（FCO）上获得了35.0％的AP，基于锚固的检测器（视网膜）获得了32.9％的AP。
Single-stage detectors suffer from extreme foreground-background class imbalance, while two-stage detectors do not. Therefore, in semi-supervised object detection, two-stage detectors can deliver remarkable performance by only selecting high-quality pseudo labels based on classification scores. However, directly applying this strategy to single-stage detectors would aggravate the class imbalance with fewer positive samples. Thus, single-stage detectors have to consider both quality and quantity of pseudo labels simultaneously. In this paper, we design a dynamic self-adaptive threshold (DSAT) strategy in classification branch, which can automatically select pseudo labels to achieve an optimal trade-off between quality and quantity. Besides, to assess the regression quality of pseudo labels in single-stage detectors, we propose a module to compute the regression uncertainty of boxes based on Non-Maximum Suppression. By leveraging only 10% labeled data from COCO, our method achieves 35.0% AP on anchor-free detector (FCOS) and 32.9% on anchor-based detector (RetinaNet).
