MixTeacher: Mining Promising Labels with Mixed Scale Teacher for Semi-Supervised Object Detection

TL;DR

MixTeacher introduces a mixed scale teacher framework for semi-supervised object detection, effectively handling scale variation by improving pseudo label quality and achieving state-of-the-art results on COCO and VOC datasets.

Contribution

The paper proposes a novel mixed scale teacher approach that enhances pseudo label generation and scale-invariant learning in semi-supervised object detection.

Findings

Achieves new state-of-the-art performance on MS COCO and PASCAL VOC.

Effectively handles extreme scale variation in semi-supervised detection.

Improves pseudo label quality through score promotion across scales.

Abstract

Scale variation across object instances remains a key challenge in object detection task. Despite the remarkable progress made by modern detection models, this challenge is particularly evident in the semi-supervised case. While existing semi-supervised object detection methods rely on strict conditions to filter high-quality pseudo labels from network predictions, we observe that objects with extreme scale tend to have low confidence, resulting in a lack of positive supervision for these objects. In this paper, we propose a novel framework that addresses the scale variation problem by introducing a mixed scale teacher to improve pseudo label generation and scale-invariant learning. Additionally, we propose mining pseudo labels using score promotion of predictions across scales, which benefits from better predictions from mixed scale features. Our extensive experiments on MS COCO and PASCAL VOC benchmarks under various semi-supervised settings demonstrate that our method achieves new state-of-the-art performance. The code and models are available at \url{https://github.com/lliuz/MixTeacher}.