Continual All-in-One Adverse Weather Removal With Knowledge Replay on a Unified Network Structure

Abstract

In real-world applications, image degeneration caused by adverse weather is always complex and changes with different weather conditions from days and seasons. Systems in real-world environments constantly encounter adverse weather conditions that are not previously observed. Therefore, it practically requires adverse weather removal models to continually learn from incrementally collected data reflecting various degeneration types. Existing adverse weather removal approaches, for either single or multiple adverse weathers, are mainly designed for a static learning paradigm, which assumes that the data of all types of degenerations to handle can be finely collected at one time before a single-phase learning process. They thus cannot directly handle the incremental learning requirements. To address this issue, we made the earliest effort to investigate the continual all-in-one adverse weather removal task, in a setting closer to real-world applications. Specifically, we develop a novel continual learning framework with effective knowledge replay (KR) on a unified network structure. Equipped with a principal component projection and an effective knowledge distillation mechanism, the proposed KR techniques are tailored for the all-in-one weather removal task. It considers the characteristics of the image restoration task with multiple degenerations in continual learning, and the knowledge for different degenerations can be shared and accumulated in the unified network structure. Extensive experimental results demonstrate the effectiveness of the proposed method to deal with this challenging task, which performs competitively to existing dedicated or joint training image restoration methods. Our code is available at https://github.com/xiaojihh/CL_all-in-one .

De Cheng

Associate Professor of the State Key Laboratory of Integrated Services Networks

De Cheng is an associate professor of the State Key Laboratory of Integrated Services Networks (ISN), Xidian University.

Yanling Ji

MEng of Artificial Intelligence

Yanling Ji is Master candidate in Northwestern Polytechnical University.

Yan Li

Laboratory Intern

Yan Li received the BE degree and the ME degree from Shandong Normal University, Jinan, China, in 2020 and 2023 respectively. Her research interests include computer vision and artificial intelligence, especially on on-device training, incremental learning and image restoration.

Dingwen Zhang

Professor of Northwestern Polytechnical University

Dingwen Zhang （张鼎文）is a Professor at BRAIN Lab, NWPU. He obtained his B.S. and Ph.D. degrees from School of Automation, Northwestern Polytechnical University (NWPU) in 2012 and 2018, respectively.