Instant-Teaching: An End-to-End Semi-Supervised Object Detection Framework

TL;DR

Instant-Teaching is an end-to-end semi-supervised object detection framework that leverages pseudo labeling and data augmentation to significantly improve detection accuracy with limited labeled data.

Contribution

It introduces a novel end-to-end SSOD framework with instant pseudo labeling and a co-rectify scheme to enhance pseudo label quality and reduce confirmation bias.

Findings

Surpasses state-of-the-art by 4.2 mAP on MS-COCO with 2% labeled data.

Outperforms existing methods by about 1.0 mAP on full MS-COCO.

Achieves over 5 mAP improvement on PASCAL VOC with limited labeled data.

Abstract

Supervised learning based object detection frameworks demand plenty of laborious manual annotations, which may not be practical in real applications. Semi-supervised object detection (SSOD) can effectively leverage unlabeled data to improve the model performance, which is of great significance for the application of object detection models. In this paper, we revisit SSOD and propose Instant-Teaching, a completely end-to-end and effective SSOD framework, which uses instant pseudo labeling with extended weak-strong data augmentations for teaching during each training iteration. To alleviate the confirmation bias problem and improve the quality of pseudo annotations, we further propose a co-rectify scheme based on Instant-Teaching, denoted as Instant-Teaching$^*$. Extensive experiments on both MS-COCO and PASCAL VOC datasets substantiate the superiority of our framework. Specifically, our method surpasses state-of-the-art methods by 4.2 mAP on MS-COCO when using $2\%$ labeled data. Even with full supervised information of MS-COCO, the proposed method still outperforms state-of-the-art methods by about 1.0 mAP. On PASCAL VOC, we can achieve more than 5 mAP improvement by applying VOC07 as labeled data and VOC12 as unlabeled data.