arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2606.31373cs.CV
收藏

通过双流双层循环优化的域自适应目标检测

Domain Adaptive Object Detection via Dual-Stream Bilevel-Cycle Optimization

School of Cyber Science and Technology, Sun Yat-sen University(中山大学网络空间安全学院);University of Chinese Academy of Sciences(中国科学院大学);School of Computer Science and Technology, Anhui University(安徽大学计算机科学与技术学院);PC Laboratory(PC实验室)

Yannan Chen, Wei Wang, Wenqiang Wang, Ruoyu Chen, Jiancheng Wang, Mingbo Yang, Yaowei Wang, Xiaochun Cao

AI总结:

提出双流双层循环优化框架,将循环自训练扩展到目标检测的回归分支,解决伪标签不可靠、训练不稳定和损失爆炸问题,在四个跨域场景取得显著效果。

摘要:

循环自训练(CST)打破了标准自训练框架的共享分类器假设,对无监督域自适应有效,并通过使用目标伪标签训练来利用未标记的目标数据。CST引入了一个目标分类器,并采用内外循环更新策略,解决了伪标签不可靠的问题,使伪标签能够跨域泛化。尽管在图像分类中取得了成功,但将CST扩展到目标检测面临三个主要挑战。首先,目标检测中CST的上限受到三种不可靠伪标签的约束,即仅分类错误、仅定位错误以及两者的组合。其次,由于目标检测涉及检测多个目标对象,直接应用CST会导致训练不稳定。第三,回归坐标的数值范围更广会导致损失爆炸。为此,我们将CST应用于分类和回归,并提出了双流双层循环优化框架。具体来说,我们在Mean Teacher上构建CST以防止训练不稳定,并使用额外的归一化将回归边界框映射到标准化空间,有效解决了损失爆炸问题。此外,我们提供了回归边界的理论推导。在四个跨域标准场景上的大量实验表明,我们的框架取得了可观的结果。

英文摘要:

Cycle self-training (CST) breaks the shared classifier assumption of the standard self-training framework, which is effective for unsupervised domain adaptation and exploits unlabeled target data by training with target pseudo-labels. CST introduces a target classifier and employs an inner-outer loop updating strategy, addressing the issue of unreliable pseudo-labels and enabling pseudo-labels to generalize across domains. Despite its success in image classification, extending CST to object detection faces three main challenges. First, the upper bound of CST in object detection is constrained by three types of unreliable pseudo-labels, such as classification error alone, localization error alone, and their combination. Second, since object detection involves detecting multiple target objects, directly applying CST leads to training insta bility. Third, a wider numerical range of regression coordinates leads to exploding losses. To this end, we apply CST to both classification and regression and propose the Dual-Stream Bilevel-Cycle Optimization framework. Specifically, we construct CST upon Mean Teacher to prevent training instability and use extra normalization to map the regression bounding box into a standardized space, effectively addressing exploding losses. Also, we provide a theoretical derivation of the regression bound. Extensive experiments across four cross domain standard scenarios demonstrate that our framework achieves considerable results.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑