RIRF: Reasoning Image Restoration Framework

TL;DR

RIRF introduces a unified image restoration framework that incorporates explicit diagnostic reasoning over degradation factors and scene semantics, leading to improved interpretability and state-of-the-art results.

Contribution

It integrates structured Chain-of-Thought reasoning into image restoration, enabling diagnosis of degradation and semantics to guide the restoration process.

Findings

Achieves state-of-the-art performance on diverse UIR benchmarks.

Provides interpretable diagnostic reasoning for image restoration.

Leverages degradation severity as reinforcement learning signals.

Abstract

Universal image restoration (UIR) aims to recover clean images from diverse and unknown degradations using a unified model. Existing UIR methods primarily focus on pixel reconstruction and often lack explicit diagnostic reasoning over degradation composition, severity, and scene semantics prior to restoration. We propose Reason and Restore (R\&R), a novel framework that integrates structured Chain-of-Thought (CoT) reasoning into the image restoration pipeline. R\&R introduces an explicit reasoner, implemented by fine-tuning Qwen3-VL, to diagnose degradation types, quantify degradation severity, infer key degradation-related factors, and describe relevant scene and object semantics. The resulting structured reasoning provides interpretable and fine-grained diagnostic priors for the restorer. To further improve restoration quality, the quantified degradation severity produced by the reasoner is leveraged as reinforcement learning (RL) signals to guide and strengthen the restorer. Unlike existing multimodal LLM-based agentic systems that decouple reasoning from low-level vision tasks, R\&R tightly couples semantic diagnostic reasoning with pixel-level restoration in a unified framework. Extensive experiments across diverse UIR benchmarks demonstrate that R\&R achieves state-of-the-art performance while offering unique interpretability into the restoration process.