NTIRE 2025 the 2nd Restore Any Image Model (RAIM) in the Wild Challenge

2025-06-02

AI Summary

This work addresses two core challenges in real-scene image restoration: joint denoising and demosaicking (JDD) under low-light conditions, and image detail enhancement/generation. We establish the first dual-task benchmark explicitly designed for realistic degradations, supporting both paired quantitative evaluation and unpaired subjective assessment. To tackle the absence of ground-truth references, we propose a novel dual-track no-reference evaluation paradigm integrating multi-scale neural networks, degradation modeling, self-supervised learning, and human visual system–guided perceptual metrics—enabling synergistic optimization of fidelity and perceptual quality. The benchmark has attracted nearly 300 participating teams, with 51 teams submitting over 600 results. State-of-the-art methods identified on this benchmark comprehensively surpass prior real-image restoration approaches, achieving new performance records across multiple metrics. These advances have received consistent endorsement from over 20 domain experts.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Large Multimodal Models (LMMs)
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Security and Privacy: Large-scale security measurements

Abstract

In this paper, we present a comprehensive overview of the NTIRE 2025 challenge on the 2nd Restore Any Image Model (RAIM) in the Wild. This challenge established a new benchmark for real-world image restoration, featuring diverse scenarios with and without reference ground truth. Participants were tasked with restoring real-captured images suffering from complex and unknown degradations, where both perceptual quality and fidelity were critically evaluated. The challenge comprised two tracks: (1) the low-light joint denoising and demosaicing (JDD) task, and (2) the image detail enhancement/generation task. Each track included two sub-tasks. The first sub-task involved paired data with available ground truth, enabling quantitative evaluation. The second sub-task dealt with real-world yet unpaired images, emphasizing restoration efficiency and subjective quality assessed through a comprehensive user study. In total, the challenge attracted nearly 300 registrations, with 51 teams submitting more than 600 results. The top-performing methods advanced the state of the art in image restoration and received unanimous recognition from all 20+ expert judges. The datasets used in Track 1 and Track 2 are available at https://drive.google.com/drive/folders/1Mgqve-yNcE26IIieI8lMIf-25VvZRs_J and https://drive.google.com/drive/folders/1UB7nnzLwqDZOwDmD9aT8J0KVg2ag4Qae, respectively. The official challenge pages for Track 1 and Track 2 can be found at https://codalab.lisn.upsaclay.fr/competitions/21334#learn_the_details and https://codalab.lisn.upsaclay.fr/competitions/21623#learn_the_details.

Problem

• Restoring real-captured images with unknown degradations
• Advancing low-light joint denoising and demosaicing techniques
• Enhancing image detail generation in unpaired real-world scenarios

Innovation

• Real-world image restoration benchmark
• Low-light joint denoising and demosaicing
• Unpaired image detail enhancement