UniCoRN: Latent Diffusion-based Unified Controllable Image Restoration Network across Multiple Degradations

2025-03-20

AI Summary

Real-world images often suffer from coupled degradations—such as blur, noise, and haze—rendering existing single-degradation restoration methods inadequate. To address this, we propose the first controllable multi-degradation restoration framework based on latent diffusion models. Our method introduces a multi-head controllable diffusion architecture coupled with a Mixture-of-Experts (MoE) adaptive control network, enabling degradation-agnostic, end-to-end training via curriculum learning—without requiring explicit degradation priors or assumptions. We further construct MetaRestore, the first multi-degradation benchmark tailored to metalens imaging. Extensive experiments demonstrate that our approach significantly outperforms both single- and multi-degradation state-of-the-art methods across multiple challenging datasets—including MetaRestore—achieving robust, high-fidelity restoration of severely compound-degraded images. This work establishes a new paradigm for unified modeling in low-level vision.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multi-instance/Multi-view Learning
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Image restoration is essential for enhancing degraded images across computer vision tasks. However, most existing methods address only a single type of degradation (e.g., blur, noise, or haze) at a time, limiting their real-world applicability where multiple degradations often occur simultaneously. In this paper, we propose UniCoRN, a unified image restoration approach capable of handling multiple degradation types simultaneously using a multi-head diffusion model. Specifically, we uncover the potential of low-level visual cues extracted from images in guiding a controllable diffusion model for real-world image restoration and we design a multi-head control network adaptable via a mixture-of-experts strategy. We train our model without any prior assumption of specific degradations, through a smartly designed curriculum learning recipe. Additionally, we also introduce MetaRestore, a metalens imaging benchmark containing images with multiple degradations and artifacts. Extensive evaluations on several challenging datasets, including our benchmark, demonstrate that our method achieves significant performance gains and can robustly restore images with severe degradations. Project page: https://codejaeger.github.io/unicorn-gh

Problem

• Handles multiple image degradation types simultaneously
• Uses a multi-head diffusion model for restoration
• Introduces MetaRestore benchmark for multi-degradation images

Innovation

• Multi-head diffusion model for unified restoration
• Low-level visual cues guide controllable diffusion
• Curriculum learning without degradation assumptions