Co-Learning: Towards Semi-Supervised Object Detection with Road-side Cameras

TL;DR

This paper introduces Co-Learning, a semi-supervised object detection framework for roadside cameras that reduces labeling costs and achieves near fully-supervised performance using only 10% labeled data.

Contribution

It proposes a novel teacher-student SSL framework with mutual learning and annotation alignment for effective semi-supervised object detection.

Findings

Achieves comparable performance to fully-supervised methods with only 10% labeled data.

Addresses pseudo-target inconsistencies and task disharmony in SSL.

Effective on edge devices like roadside cameras.

Abstract

Recently, deep learning has experienced rapid expansion, contributing significantly to the progress of supervised learning methodologies. However, acquiring labeled data in real-world settings can be costly, labor-intensive, and sometimes scarce. This challenge inhibits the extensive use of neural networks for practical tasks due to the impractical nature of labeling vast datasets for every individual application. To tackle this, semi-supervised learning (SSL) offers a promising solution by using both labeled and unlabeled data to train object detectors, potentially enhancing detection efficacy and reducing annotation costs. Nevertheless, SSL faces several challenges, including pseudo-target inconsistencies, disharmony between classification and regression tasks, and efficient use of abundant unlabeled data, especially on edge devices, such as roadside cameras. Thus, we developed a teacher-student-based SSL framework, Co-Learning, which employs mutual learning and annotation-alignment strategies to adeptly navigate these complexities and achieves comparable performance as fully-supervised solutions using 10\% labeled data.