CRTrack: Low-Light Semi-Supervised Multi-object Tracking Based on Consistency Regularization

2025-02-24

AI Summary

Low-light multi-object tracking (MOT) has long been hindered by high annotation costs and scarce labeled data. To address this, we introduce LLMOT—the first dedicated low-light MOT benchmark—and propose CRTrack, a semi-supervised framework. CRTrack features: (1) consistency regularization coupled with low-light image enhancement to improve feature robustness; (2) an adaptive sampling assignment mechanism that replaces static IoU-based matching to mitigate detection bias under poor illumination; and (3) a dynamic network updating strategy to refine pseudo-label quality. Evaluated on LLMOT, CRTrack achieves state-of-the-art fully supervised performance using only 10% of annotated data, substantially reducing annotation dependency. Both the LLMOT dataset and CRTrack code are publicly released.

Technology Category

• Computer Vision: Motion & Tracking
• Machine Learning: Semi-Supervised Learning
• Search and Optimization: Learning to Search

Application Category

• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Multi-object tracking under low-light environments is prevalent in real life. Recent years have seen rapid development in the field of multi-object tracking. However, due to the lack of datasets and the high cost of annotations, multi-object tracking under low-light environments remains a persistent challenge. In this paper, we focus on multi-object tracking under low-light conditions. To address the issues of limited data and the lack of dataset, we first constructed a low-light multi-object tracking dataset (LLMOT). This dataset comprises data from MOT17 that has been enhanced for nighttime conditions as well as multiple unannotated low-light videos. Subsequently, to tackle the high annotation costs and address the issue of image quality degradation, we propose a semi-supervised multi-object tracking method based on consistency regularization named CRTrack. First, we calibrate a consistent adaptive sampling assignment to replace the static IoU-based strategy, enabling the semi-supervised tracking method to resist noisy pseudo-bounding boxes. Then, we design a adaptive semi-supervised network update method, which effectively leverages unannotated data to enhance model performance. Dataset and Code: https://github.com/ZJZhao123/CRTrack.

Problem

• Addresses multi-object tracking in low-light conditions
• Reduces high annotation costs with semi-supervised method
• Improves image quality degradation in low-light environments

Innovation

• Constructed low-light multi-object tracking dataset
• Proposed consistency regularization-based tracking method
• Designed adaptive semi-supervised network update