Low-Confidence Samples Mining for Semi-supervised Object Detection

TL;DR

This paper introduces a Low-confidence Samples Mining method that leverages low-confidence pseudo-labels in semi-supervised object detection, significantly improving performance by extracting valuable information from less confident predictions.

Contribution

The paper proposes a novel LSM approach with a PIM branch and self-distillation, enabling effective use of low-confidence pseudo-labels in SSOD, applicable to different detection architectures.

Findings

Achieves 3.54% mAP improvement on MS-COCO with 5% labels

Effectively utilizes low-confidence pseudo-labels for better detection

Compatible with Faster-RCNN and Deformable-DETR architectures

Abstract

Reliable pseudo-labels from unlabeled data play a key role in semi-supervised object detection (SSOD). However, the state-of-the-art SSOD methods all rely on pseudo-labels with high confidence, which ignore valuable pseudo-labels with lower confidence. Additionally, the insufficient excavation for unlabeled data results in an excessively low recall rate thus hurting the network training. In this paper, we propose a novel Low-confidence Samples Mining (LSM) method to utilize low-confidence pseudo-labels efficiently. Specifically, we develop an additional pseudo information mining (PIM) branch on account of low-resolution feature maps to extract reliable large-area instances, the IoUs of which are higher than small-area ones. Owing to the complementary predictions between PIM and the main branch, we further design self-distillation (SD) to compensate for both in a mutually-learning manner. Meanwhile, the extensibility of the above approaches enables our LSM to apply to Faster-RCNN and Deformable-DETR respectively. On the MS-COCO benchmark, our method achieves 3.54% mAP improvement over state-of-the-art methods under 5% labeling ratios.