While domain adaptation has been used to improve the performance of object detectors when the training and test data follow different distributions, previous work has mostly focused on two-stage detectors. This is because their use of region proposals makes it possible to perform local adaptation, which has been shown to significantly improve the adaptation effectiveness. Here, by contrast, we target single-stage architectures, which are better suited to resource-constrained detection than two-stage ones but do not provide region proposals. To nonetheless benefit from the strength of local adaptation, we introduce an attention mechanism that lets us identify the important regions on which adaptation should focus. Our method gradually adapts the features from global, image-level to local, instance-level. Our approach is generic and can be integrated into any single-stage detector. We demonstrate this on standard benchmark datasets by applying it to both SSD and YOLOv5. Furthermore, for equivalent single-stage architectures, our method outperforms the state-of-the-art domain adaptation techniques even though they were designed for specific detectors.
翻译:虽然当培训和测试数据分布不同时,已利用领域适应来改进物体探测器的性能,但先前的工作主要侧重于两阶段探测器,这是因为使用区域建议可以进行地方性适应,这证明大大提高了适应效果。与此形成对照,我们的目标是单阶段结构,这些结构比两阶段更适合资源紧张的探测,但并不提供区域建议。然而,为了从地方性适应的力量中获益,我们引入了一个关注机制,让我们能够确定适应工作应当侧重的重要区域。我们的方法逐渐将全球图像水平的特征从全球图像水平调整到地方实例水平。我们的方法是通用的,可以纳入任何单阶段的探测器。我们在标准基准数据集上展示这一点,将它应用到SSD和YOLOv5.此外,对于同等的单阶段结构,我们的方法超越了为特定探测器设计的最先进的领域适应技术。