Learning Domain Adaptive Object Detection with Probabilistic Teacher
This work addresses the problem of domain adaptation in object detection for computer vision applications, offering an incremental improvement by focusing on uncertainty estimation in pseudo-boxes.
The paper tackles the challenge of improving pseudo-box quality in self-training for unsupervised domain adaptive object detection by introducing the Probabilistic Teacher framework, which leverages uncertainty-guided consistency training and a novel Entropy Focal Loss to achieve new state-of-the-art results with significant performance gains.
Self-training for unsupervised domain adaptive object detection is a challenging task, of which the performance depends heavily on the quality of pseudo boxes. Despite the promising results, prior works have largely overlooked the uncertainty of pseudo boxes during self-training. In this paper, we present a simple yet effective framework, termed as Probabilistic Teacher (PT), which aims to capture the uncertainty of unlabeled target data from a gradually evolving teacher and guides the learning of a student in a mutually beneficial manner. Specifically, we propose to leverage the uncertainty-guided consistency training to promote classification adaptation and localization adaptation, rather than filtering pseudo boxes via an elaborate confidence threshold. In addition, we conduct anchor adaptation in parallel with localization adaptation, since anchor can be regarded as a learnable parameter. Together with this framework, we also present a novel Entropy Focal Loss (EFL) to further facilitate the uncertainty-guided self-training. Equipped with EFL, PT outperforms all previous baselines by a large margin and achieve new state-of-the-arts.