AIR-DA: Adversarial Image Reconstruction for Unsupervised Domain Adaptive Object Detection
This work addresses domain adaptation for object detection, which is crucial for deploying models in real-world scenarios with varying data distributions, but it is incremental as it builds on existing adversarial alignment methods.
The paper tackles the problem of domain shift in unsupervised domain adaptive object detection by introducing an auxiliary regularization task to improve adversarial training balance, achieving state-of-the-art performance across multiple datasets.
Unsupervised domain adaptive object detection is a challenging vision task where object detectors are adapted from a label-rich source domain to an unlabeled target domain. Recent advances prove the efficacy of the adversarial based domain alignment where the adversarial training between the feature extractor and domain discriminator results in domain-invariance in the feature space. However, due to the domain shift, domain discrimination, especially on low-level features, is an easy task. This results in an imbalance of the adversarial training between the domain discriminator and the feature extractor. In this work, we achieve a better domain alignment by introducing an auxiliary regularization task to improve the training balance. Specifically, we propose Adversarial Image Reconstruction (AIR) as the regularizer to facilitate the adversarial training of the feature extractor. We further design a multi-level feature alignment module to enhance the adaptation performance. Our evaluations across several datasets of challenging domain shifts demonstrate that the proposed method outperforms all previous methods, of both one- and two-stage, in most settings.