论文标题
SSDA-YOLO：半监督域自适应YOLO用于跨域对象检测
SSDA-YOLO: Semi-supervised Domain Adaptive YOLO for Cross-Domain Object Detection
论文作者
论文摘要
域自适应对象检测（DAOD）旨在减轻跨域差异引起的转移性能降解。但是，大多数现有的daod方法都以过时的和计算密集的两阶段更快的速度主导，这不是工业应用的首选。在本文中，我们提出了一种基于新型的半监督域自适应Yolo（SSDA-Yolo）方法，以通过将紧凑的一阶段较强的较强的检测器Yolov5与域适应性整合在一起，以提高跨域检测性能。具体来说，我们将知识蒸馏框架与平均教师模型一起调整，以帮助学生模型获得未标记的目标域的实例级特征。我们还利用场景样式转移到不同域中的交叉生成伪图像，以纠正图像级别的差异。另外，提出了直观的一致性损失，以进一步对齐跨域预测。我们通过公共基准评估SSDA-YOLO，包括Pascalvoc，Clipart1k，CityScapes和Foggy CityScapes。此外，为了验证其概括，我们对从各个真实教室收集的打哈欠检测数据集进行了实验。结果表明，在这些daod任务中，我们的方法有了很大的改善，这既揭示了提出的自适应模块的有效性，又揭示了在daod中应用更多高级检测器的紧迫性。我们的代码可在\ url {https://github.com/hhnuzhy/ssda-yolo}上找到。
Domain adaptive object detection (DAOD) aims to alleviate transfer performance degradation caused by the cross-domain discrepancy. However, most existing DAOD methods are dominated by outdated and computationally intensive two-stage Faster R-CNN, which is not the first choice for industrial applications. In this paper, we propose a novel semi-supervised domain adaptive YOLO (SSDA-YOLO) based method to improve cross-domain detection performance by integrating the compact one-stage stronger detector YOLOv5 with domain adaptation. Specifically, we adapt the knowledge distillation framework with the Mean Teacher model to assist the student model in obtaining instance-level features of the unlabeled target domain. We also utilize the scene style transfer to cross-generate pseudo images in different domains for remedying image-level differences. In addition, an intuitive consistency loss is proposed to further align cross-domain predictions. We evaluate SSDA-YOLO on public benchmarks including PascalVOC, Clipart1k, Cityscapes, and Foggy Cityscapes. Moreover, to verify its generalization, we conduct experiments on yawning detection datasets collected from various real classrooms. The results show considerable improvements of our method in these DAOD tasks, which reveals both the effectiveness of proposed adaptive modules and the urgency of applying more advanced detectors in DAOD. Our code is available on \url{https://github.com/hnuzhy/SSDA-YOLO}.
