MUM : Mix Image Tiles and UnMix Feature Tiles for Semi-Supervised Object Detection

TL;DR

This paper introduces MUM, a novel data augmentation technique for semi-supervised object detection that unmixes feature tiles to create effective weak-strong input pairs, improving detection performance.

Contribution

MUM is a simple, effective augmentation method that reconstructs mixed image tiles in feature space, compatible with various SSOD methods and enhancing detection accuracy.

Findings

Consistently improves mAP across benchmarks

Effective in creating meaningful weak-strong pairs

Enhances SSOD performance on MS-COCO and PASCAL VOC

Abstract

Many recent semi-supervised learning (SSL) studies build teacher-student architecture and train the student network by the generated supervisory signal from the teacher. Data augmentation strategy plays a significant role in the SSL framework since it is hard to create a weak-strong augmented input pair without losing label information. Especially when extending SSL to semi-supervised object detection (SSOD), many strong augmentation methodologies related to image geometry and interpolation-regularization are hard to utilize since they possibly hurt the location information of the bounding box in the object detection task. To address this, we introduce a simple yet effective data augmentation method, Mix/UnMix (MUM), which unmixes feature tiles for the mixed image tiles for the SSOD framework. Our proposed method makes mixed input image tiles and reconstructs them in the feature space. Thus, MUM can enjoy the interpolation-regularization effect from non-interpolated pseudo-labels and successfully generate a meaningful weak-strong pair. Furthermore, MUM can be easily equipped on top of various SSOD methods. Extensive experiments on MS-COCO and PASCAL VOC datasets demonstrate the superiority of MUM by consistently improving the mAP performance over the baseline in all the tested SSOD benchmark protocols.