Unbiased Teacher for Semi-Supervised Object Detection

TL;DR

This paper introduces Unbiased Teacher, a novel semi-supervised object detection method that mitigates pseudo-label bias, significantly improving detection performance with limited labeled data.

Contribution

The paper proposes Unbiased Teacher, a new approach that jointly trains a student and a gradually progressing teacher, incorporating class-balance loss to enhance semi-supervised object detection.

Findings

Achieves 6.8 mAP improvement with 1% labeled data on MS-COCO.

Outperforms supervised baseline by around 10 mAP with minimal labeled data.

Consistently improves state-of-the-art methods across multiple datasets.

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.