Cycle Self-Training for Semi-Supervised Object Detection with Distribution Consistency Reweighting

Abstract

Recently, many semi-supervised object detection (SSOD) methods adopt teacher-student framework and have achieved state-of-the-art results. However, the teacher network is tightly coupled with the student network since the teacher is an exponential moving average (EMA) of the student, which causes a performance bottleneck. To address the coupling problem, we propose a Cycle Self-Training (CST) framework for SSOD, which consists of two teachers T1 and T2, two students S1 and S2. Based on these networks, a cycle self-training mechanism is built, i.e., S1$\to$T1$\to$S2$\to$T2$\to$S1. For S$\to$T, we also utilize the EMA weights of the students to update the teachers. For T$\to$S, instead of providing supervision for its own student S1(S2) directly, the teacher T1(T2) generates pseudo-labels for the student S2(S1), which looses the coupling effect. Moreover, owing to the property of EMA, the teacher is most likely to accumulate the biases from the student and make the mistakes irreversible. To mitigate the problem, we also propose a distribution consistency reweighting strategy, where pseudo-labels are reweighted based on distribution consistency across the teachers T1 and T2. With the strategy, the two students S2 and S1 can be trained robustly with noisy pseudo labels to avoid confirmation biases. Extensive experiments prove the superiority of CST by consistently improving the AP over the baseline and outperforming state-of-the-art methods by 2.1% absolute AP improvements with scarce labeled data.

arXiv commentACM Multimedia 2022

References in corpus (10)

• YOLOv4: Optimal Speed and Accuracy of Object Detection
• mixup: Beyond Empirical Risk Minimization
• FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence
• Deformable DETR: Deformable Transformers for End-to-End Object Detection
• Unsupervised Data Augmentation for Consistency Training
• Weighted boxes fusion: Ensembling boxes from different object detection models
• FlexMatch: Boosting Semi-Supervised Learning with Curriculum Pseudo Labeling
• A Simple Semi-Supervised Learning Framework for Object Detection
• Combating Noise: Semi-supervised Learning by Region Uncertainty Quantification
• Rethinking Pseudo Labels for Semi-Supervised Object Detection