Reshaping the Online Data Buffering and Organizing Mechanism for Continual Test-Time Adaptation

TL;DR

This paper introduces a novel buffering and organizing mechanism for Continual Test-Time Adaptation, improving model robustness against domain shifts by using uncertainty-aware sample selection, class relation preservation, and pseudo-target replay.

Contribution

It proposes an uncertainty-aware buffering method, a graph-based class relation constraint, and a pseudo-target replay objective to enhance CTTA performance.

Findings

Outperforms existing methods in segmentation and classification CTTA tasks.

Effectively mitigates error accumulation and catastrophic forgetting.

Demonstrates robustness across diverse domain shifts.

Abstract

Continual Test-Time Adaptation (CTTA) involves adapting a pre-trained source model to continually changing unsupervised target domains. In this paper, we systematically analyze the challenges of this task: online environment, unsupervised nature, and the risks of error accumulation and catastrophic forgetting under continual domain shifts. To address these challenges, we reshape the online data buffering and organizing mechanism for CTTA. We propose an uncertainty-aware buffering approach to identify and aggregate significant samples with high certainty from the unsupervised, single-pass data stream. Based on this, we propose a graph-based class relation preservation constraint to overcome catastrophic forgetting. Furthermore, a pseudo-target replay objective is used to mitigate error accumulation. Extensive experiments demonstrate the superiority of our method in both segmentation and classification CTTA tasks. Code is available at https://github.com/z1358/OBAO.