重新思考航空图像半监督目标检测中的尺度不平衡问题
Rethinking Scale Imbalance in Semi-supervised Object Detection for Aerial Images
AI总结：
本文针对航空图像半监督目标检测中的尺度不平衡问题，提出S^3OD学习流程，通过SAT、SLA和TNL三个组件解决伪标签、标签分配和负学习的不平衡，在DOTA-v1.5上表现优于现有方法。
摘要：
本文关注航空图像中半监督目标检测（SSOD）的尺度不平衡问题。与自然图像相比，航空图像中的目标尺寸更小且每张图像中的数量更多，增加了人工标注的难度。同时，先进的SSOD技术能够利用有限的标注数据和大量未标注数据训练出优秀的检测器，从而节省标注成本。然而，作为航空图像中研究不足的任务，SSOD在面对大量小目标时会出现性能急剧下降。通过分析小目标和大目标之间的预测结果，我们识别出由尺度偏差引起的三个不平衡问题，即伪标签不平衡、标签分配不平衡和负学习不平衡。为了解决这些问题，我们提出了一种用于航空图像的新型尺度判别半监督目标检测（S^3OD）学习流程。在我们的S^3OD中，提出了三个关键组件：尺寸感知自适应阈值（SAT）、尺寸重平衡标签分配（SLA）和教师引导负学习（TNL），以确保尺度无偏学习。具体而言，SAT自适应地选择合适的阈值来过滤不同尺度目标的伪标签。SLA通过重采样和重加权来平衡不同尺度目标的正样本。TNL利用教师模型生成的信息来缓解负样本中的不平衡。在DOTA-v1.5基准上进行的大量实验证明了我们提出方法相对于最先进竞争对手的优越性。代码即将发布。
英文摘要：
This paper focuses on the scale imbalance problem of semi-supervised object detection(SSOD) in aerial images. Compared to natural images, objects in aerial images show smaller sizes and larger quantities per image, increasing the difficulty of manual annotation. Meanwhile, the advanced SSOD technique can train superior detectors by leveraging limited labeled data and massive unlabeled data, saving annotation costs. However, as an understudied task in aerial images, SSOD suffers from a drastic performance drop when facing a large proportion of small objects. By analyzing the predictions between small and large objects, we identify three imbalance issues caused by the scale bias, i.e., pseudo-label imbalance, label assignment imbalance, and negative learning imbalance. To tackle these issues, we propose a novel Scale-discriminative Semi-Supervised Object Detection (S^3OD) learning pipeline for aerial images. In our S^3OD, three key components, Size-aware Adaptive Thresholding (SAT), Size-rebalanced Label Assignment (SLA), and Teacher-guided Negative Learning (TNL), are proposed to warrant scale unbiased learning. Specifically, SAT adaptively selects appropriate thresholds to filter pseudo-labels for objects at different scales. SLA balances positive samples of objects at different scales through resampling and reweighting. TNL alleviates the imbalance in negative samples by leveraging information generated by a teacher model. Extensive experiments conducted on the DOTA-v1.5 benchmark demonstrate the superiority of our proposed methods over state-of-the-art competitors. Codes will be released soon.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
