Recently,the detection transformer has gained substantial attention for its inherent minimal post-processing requirement.However,this paradigm relies on abundant training data,yet in the context of the cross-domain adaptation,insufficient labels in the target domain exacerbate issues of class imbalance and model performance degradation.To address these challenges, we propose a novel class-aware cross domain detection transformer based on the adversarial learning and mean-teacher framework.First,considering the inconsistencies between the classification and regression tasks,we introduce an IoU-aware prediction branch and exploit the consistency of classification and location scores to filter and reweight pseudo labels.Second, we devise a dynamic category threshold refinement to adaptively manage model confidence.Third,to alleviate the class imbalance,an instance-level class-aware contrastive learning module is presented to encourage the generation of discriminative features for each class,particularly benefiting minority classes.Experimental results across diverse domain-adaptive scenarios validate our method's effectiveness in improving performance and alleviating class imbalance issues,which outperforms the state-of-the-art transformer based methods.

翻译：最近，检测Transformer因其固有的最小后处理需求而受到广泛关注。然而，这一范式依赖于充足的训练数据，但在跨域适应背景下，目标域中标签不足会加剧类别不平衡和模型性能退化的问题。为应对这些挑战，我们提出了一种基于对抗学习和均值教师框架的新型类别感知跨域检测Transformer。首先，考虑到分类与回归任务之间的不一致性，我们引入了一个IoU感知预测分支，并利用分类与定位分数的一致性来过滤和重新加权伪标签。其次，我们设计了一种动态类别阈值细化方法，以自适应地管理模型置信度。第三，为缓解类别不平衡，我们提出了一个实例级类别感知对比学习模块，鼓励为每个类别生成判别性特征，尤其有益于少数类别。在多种域适应场景下的实验结果验证了我们的方法在提升性能和缓解类别不平衡问题上的有效性，其表现超越了当前最优的基于Transformer的方法。