Recent advancements in all-in-one image restoration models have revolutionized the ability to address diverse degradations through a unified framework. However, parameters tied to specific tasks often remain inactive for other tasks, making mixture-of-experts (MoE) architectures a natural extension. Despite this, MoEs often show inconsistent behavior, with some experts unexpectedly generalizing across tasks while others struggle within their intended scope. This hinders leveraging MoEs' computational benefits by bypassing irrelevant experts during inference. We attribute this undesired behavior to the uniform and rigid architecture of traditional MoEs. To address this, we introduce ``complexity experts" -- flexible expert blocks with varying computational complexity and receptive fields. A key challenge is assigning tasks to each expert, as degradation complexity is unknown in advance. Thus, we execute tasks with a simple bias toward lower complexity. To our surprise, this preference effectively drives task-specific allocation, assigning tasks to experts with the appropriate complexity. Extensive experiments validate our approach, demonstrating the ability to bypass irrelevant experts during inference while maintaining superior performance. The proposed MoCE-IR model outperforms state-of-the-art methods, affirming its efficiency and practical applicability.
(a) Dense all-in-one restoration methods often inefficiently allocate parameters when handling multiple degradation types.
(b) While recent Mixture-of-Experts (MoE) approaches address this through sparse computation, their rigid routing mechanisms uniformly distribute inputs across experts without considering the natural relationships between degradations.
(c) To overcome these limitations, we introduce Complexity Experts - adaptive processing blocks with size-varying computational units. Our framework dynamically allocates model capacity using a spring-inspired force mechanism that continuously guides routing decisions toward simpler experts when possible, with the force proportional to the complexity of the input degradation. While initially designed for computational efficiency, this approach naturally emerges as a task-discriminative learning framework, assigning degradations to the most suitable experts. This makes it particularly effective for all-in-one restoration methods, where both task-specific processing and cross-degradation knowledge sharing are crucial.
@misc{zamfir2024complexityexperts,
title={Complexity Experts are Task-Discriminative Learners for Any Image Restoration},
author={Eduard Zamfir and Zongwei Wu and Nancy Mehta and Yuedong Tan and Danda Pani Paudel and Yulun Zhang and Radu Timofte},
year={2024},
eprint={2411.18466},
archivePrefix={arXiv},
primaryClass={cs.CV},
}