Feature Alignment and Uniformity for Test Time Adaptation

Shuai Wang1; Daoan Zhang2; Zipei Yan3; Jianguo Zhang2; Rui Li1

10.1109/CVPR52729.2023.01920

2023

1063-6919

979-8-3503-0130-4

2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

2023-June

20050-20060

17-24 June 2023

Vancouver, BC, Canada

Test time adaptation (TTA) aims to adapt deep neural networks when receiving out of distribution test domain samples. In this setting, the model can only access online unlabeled test samples and pretrained models on the training domains. We first address TTA as a feature revision problem due to the domain gap between source domains and target domains. After that, we follow the two measurements alignment and uniformity to discuss the test time feature revision. For test time feature uniformity, we propose a test time self-distillation strategy to guarantee the consistency of uniformity between representations of the current batch and all the previous batches. For test time feature alignment, we propose a memorized spatial local clustering strategy to align the representations among the neighborhood samples for the upcoming batch. To deal with the common noisy label problem, we propound the entropy and consistency filters to select and drop the possible noisy labels. To prove the scalability and efficacy of our method, we conduct experiments on four domain generalization bench marks and four medical image segmentation tasks with various backbones. Experiment results show that our method not only improves baseline stably but also outperforms existing state-of-the-art test time adaptation methods.

Transfer meta low-shot continual or long-tail learning

其他

CPCI-S ; EI

WOS:001062531304036

20234114867428

IEEE

1.Tsinghua University
2.Southern University of Science and Technology
3.Hongkong Polytechnic University

Shuai Wang,Daoan Zhang,Zipei Yan,et al. Feature Alignment and Uniformity for Test Time Adaptation[C],2023:20050-20060.