领域自适应目标检测的对比均值教师（英文版）

英文标题：Contrastive Mean Teacher for Domain Adaptive Object Detectors中文摘要：本文提出了一种统一的、通用的框架 ——CMT，将 mean-teacher 自训练和对比学习这两种范式自然地集成在一起，通过伪标签进行特征抽取和优化，从而稳定提高了目标域上的性能，取得了新的最优表现。英文摘要：Object detectors often suffer from the domain gap between training (sourcedomain) and real-world applications (target domain). Mean-teacher self-trainingis a powerful paradigm in unsupervised domain adaptation for object detection,but it struggles with low-quality pseudo-labels. In this work, we identify theintriguing alig