利用目标图像中模型自信区域增强检测器，无需源数据即可适应新场景。
Leveraging Confident Image Regions for Source-Free Domain-Adaptive Object Detection
- 从目标图像中裁剪模型自信区域进行增强
- 在两个交通场景基准上达到新最好性能
- 适合无源数据时的域自适应检测任务
源域无关的域自适应目标检测是一个有趣但研究较少的问题，旨在不使用源数据的情况下，将预训练于源域的检测器适配到目标域。目前尚无专为该任务设计的数据增强方法。本文提出一种新型数据增强策略：裁剪目标图像中检测器置信度高的区域，连同其伪标签进行增强，并拼接成更具挑战性的目标图像以优化检测器。由于源数据在适配过程中不可用，采用教师-学生学习框架防止模型崩溃。在三个交通场景适配基准上评估，其中两个达到新的最先进水平。
原文摘要 · Abstract (English)
Source-free domain-adaptive object detection is an interesting but scarcely addressed topic. It aims at adapting a source-pretrained detector to a distinct target domain without resorting to source data during adaptation. So far, there is no data augmentation scheme tailored to source-free domain-adaptive object detection. To this end, this paper presents a novel data augmentation approach that cuts out target image regions where the detector is confident, augments them along with their respective pseudo-labels, and joins them into a challenging target image to adapt the detector. As the source data is out of reach during adaptation, we implement our approach within a teacher-student learning paradigm to ensure that the model does not collapse during the adaptation procedure. We evaluated our approach on three adaptation benchmarks of traffic scenes, scoring new state-of-the-art on two of them.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。