CLDA-YOLO: Visual Contrastive Learning Based Domain Adaptive YOLO Detector

TL;DR

This paper introduces CLDA-YOLO, a novel unsupervised domain adaptive object detector based on visual contrastive learning, specifically designed for YOLO, which improves cross-domain robustness without sacrificing inference speed.

Contribution

It develops a comprehensive domain adaptive architecture for YOLO using a teacher-student system, uncertainty learning, dynamic data augmentation, and a unified contrastive learning paradigm.

Findings

Achieves highly competitive results across multiple domain datasets.

Maintains inference speed while improving domain robustness.

Effectively aligns features at backbone and head stages.

Abstract

Unsupervised domain adaptive (UDA) algorithms can markedly enhance the performance of object detectors under conditions of domain shifts, thereby reducing the necessity for extensive labeling and retraining. Current domain adaptive object detection algorithms primarily cater to two-stage detectors, which tend to offer minimal improvements when directly applied to single-stage detectors such as YOLO. Intending to benefit the YOLO detector from UDA, we build a comprehensive domain adaptive architecture using a teacher-student cooperative system for the YOLO detector. In this process, we propose uncertainty learning to cope with pseudo-labeling generated by the teacher model with extreme uncertainty and leverage dynamic data augmentation to asymptotically adapt the teacher-student system to the environment. To address the inability of single-stage object detectors to align at multiple stages, we utilize a unified visual contrastive learning paradigm that aligns instance at backbone and head respectively, which steadily improves the robustness of the detectors in cross-domain tasks. In summary, we present an unsupervised domain adaptive YOLO detector based on visual contrastive learning (CLDA-YOLO), which achieves highly competitive results across multiple domain adaptive datasets without any reduction in inference speed.