UnSCAR: Universal, Scalable, Controllable, and Adaptable Image Restoration

TL;DR

This work introduces a unified inference pipeline with a multi-branch mixture-of-experts architecture that decomposes restoration knowledge across specialized task-adaptable experts that enables scalable learning, adapts and generalizes robustly to unseen domains, and supports user-controllable restoration across degradations.

Abstract

Universal image restoration aims to recover clean images from arbitrary real-world degradations using a single inference model. Despite significant progress, existing all-in-one restoration networks do not scale to multiple degradations. As the number of degradations increases, training becomes unstable, models grow excessively large, and performance drops across both seen and unseen domains. In this work, we show that scaling universal restoration is fundamentally limited by interference across degradations during joint learning, leading to catastrophic task forgetting. To address this challenge, we introduce a unified inference pipeline with a multi-branch mixture-of-experts architecture that decomposes restoration knowledge across specialized task-adaptable experts. Our approach enables scalable learning (over sixteen degradations), adapts and generalizes robustly to unseen domains, and supports user-controllable restoration across degradations. Beyond achieving superior performance across benchmarks, this work establishes a new design paradigm for scalable and controllable universal image restoration.

Figures (12)

• Figure 1: We introduce UnSCAR, a universal image restoration model that can handle $2\times$ more degradations than previously explored, achieving state-of-the-art performance across multiple benchmarks. Our framework also enables user-controllable restoration via degradation-control sliders and supports few-shot adaptation (down to one-shot) with strong out-of-distribution generalization on challenging medical imaging datasets.
• Figure 2: Low-level cues. (a) Low-level cues provide discriminative signals across degradation types, enabling automatic restoration. (b) Our framework leverages diverse cues such as edges, color statistics, noise, and transmission, to characterize the input scene.
• Figure 3: UnSCAR architecture. We leverage low-level visual guidance within a bidirectional feedback, controllable generative network with mixture-of-experts. This enables unified image restoration across multiple simultaneous degradations.
• Figure 4: Degradation-aware embeddings. Our degradation-aware training framework (left) learns an auxilliary degradation encoder cloned from the CLIP image encoder trained with SigLIP loss zhai2023sigmoid. The resulting embeddings provide clear separation between single- and mixed-degradation tasks while maintaining proximity across related degradations.
• Figure 5: Single-degradation restoration.UnSCAR produces sharper, higher-fidelity reconstructions across single-degradation tasks.