Unbiased Teacher for Semi-Supervised Object Detection

Abstract

Semi-supervised learning, i.e., training networks with both labeled and unlabeled data, has made significant progress recently. However, existing works have primarily focused on image classification tasks and neglected object detection which requires more annotation effort. In this work, we revisit the Semi-Supervised Object Detection (SS-OD) and identify the pseudo-labeling bias issue in SS-OD. To address this, we introduce Unbiased Teacher, a simple yet effective approach that jointly trains a student and a gradually progressing teacher in a mutually-beneficial manner. Together with a class-balance loss to downweight overly confident pseudo-labels, Unbiased Teacher consistently improved state-of-the-art methods by significant margins on COCO-standard, COCO-additional, and VOC datasets. Specifically, Unbiased Teacher achieves 6.8 absolute mAP improvements against state-of-the-art method when using 1% of labeled data on MS-COCO, achieves around 10 mAP improvements against the supervised baseline when using only 0.5, 1, 2% of labeled data on MS-COCO.

Accepted to ICLR 2021; Code is available at https://github.com/facebookresearch/unbiased-teacher

References in corpus (7)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• YOLOv3: An Incremental Improvement
• Bootstrap your own latent: A new approach to self-supervised Learning
• Improved Regularization of Convolutional Neural Networks with Cutout
• Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results
• A Simple Semi-Supervised Learning Framework for Object Detection
• LSDA: Large Scale Detection Through Adaptation

Cited by in corpus (7)

• Localizing Objects with Self-Supervised Transformers and no Labels
• SODA10M: A Large-Scale 2D Self/Semi-Supervised Object Detection Dataset for Autonomous Driving
• Decoupled Adaptation for Cross-Domain Object Detection
• Uncertainty-aware Mean Teacher for Source-free Unsupervised Domain Adaptive 3D Object Detection
• Combating Noise: Semi-supervised Learning by Region Uncertainty Quantification
• Structured Sparse R-CNN for Direct Scene Graph Generation
• Hetero-Modal Learning and Expansive Consistency Constraints for Semi-Supervised Detection from Multi-Sequence Data