arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新，休息一下吧，周末愉快~~
arXiv 2301.04795cs.CV
收藏

ECCV 2022 OOD-CV挑战赛图像分类赛道第一名解决方案

1st Place Solution for ECCV 2022 OOD-CV Challenge Image Classification Track

Hikvision Research Institute（海康威视研究院）；Peking University Shenzhen Graduate School（北京大学深圳研究生院）；Zhejiang University（浙江大学）

Yilu Guo, Xingyue Shi, Weijie Chen, Shicai Yang, Di Xie, Shiliang Pu, Yueting Zhuang

备注：
Tech Report
阅读 PDF
arXiv 页面 ↗HTML 原文 ↗

摘要：

OOD-CV挑战是一个分布外泛化任务。在这个挑战中，我们的核心解决方案可以概括为，噪声标签学习是一个强大的测试时领域自适应优化器。简而言之，我们的主要管道可以分为两个阶段，一个预训练阶段的领域泛化和领域自适应的测试时训练阶段。我们只利用标记的源数据在预训练阶段，只利用未标记的目标数据在测试时训练阶段。在预训练阶段，我们提出了一种简单而有效的掩码级复制粘贴数据增强策略，以增强分布外泛化能力，从而抵抗形状，姿态，上下文，纹理，遮挡和天气领域的变化。在测试时训练阶段，使用预训练好的模型为未标记的目标数据分配噪声标签，并提出了一种用于噪声标签学习的Label-Periodically-Updated DivideMix方法。在集成了测试时增强和模型包围策略后，我们的解决方案在OOD-CV挑战赛的图像分类排行榜上名列第一。代码将在https://github.com/hikvision-research/OOD-CV上发布。

英文摘要：

OOD-CV challenge is an out-of-distribution generalization task. In this challenge, our core solution can be summarized as that Noisy Label Learning Is A Strong Test-Time Domain Adaptation Optimizer. Briefly speaking, our main pipeline can be divided into two stages, a pre-training stage for domain generalization and a test-time training stage for domain adaptation. We only exploit labeled source data in the pre-training stage and only exploit unlabeled target data in the test-time training stage. In the pre-training stage, we propose a simple yet effective Mask-Level Copy-Paste data augmentation strategy to enhance out-of-distribution generalization ability so as to resist shape, pose, context, texture, occlusion, and weather domain shifts in this challenge. In the test-time training stage, we use the pre-trained model to assign noisy label for the unlabeled target data, and propose a Label-Periodically-Updated DivideMix method for noisy label learning. After integrating Test-Time Augmentation and Model Ensemble strategies, our solution ranks the first place on the Image Classification Leaderboard of the OOD-CV Challenge. Code will be released in https://github.com/hikvision-research/OOD-CV.

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。

↑