AIM 2025 Challenge on Real-World RAW Image Denoising

2025-10-07

AI Summary

To address poor generalization of low-light RAW image denoising across real-world multi-camera systems, this paper introduces Multi-Cam-RAW—the first cross-device low-light RAW noise benchmark—and establishes camera-agnostic RAW-domain denoising as a new challenge. Methodologically, we design a controllable, physically grounded noise synthesis pipeline to generate high-fidelity paired data for training a lightweight denoiser; we further propose a hybrid evaluation framework integrating full-reference metrics (PSNR, SSIM, LPIPS) and no-reference metrics (ARNIQA, TOPIQ). Our key contributions are: (1) the first synthetic–real hybrid benchmark tailored for multi-camera RAW denoising; (2) empirical validation that joint optimization of noise modeling and network architecture is critical for cross-device generalization; and (3) advancement of robust image restoration techniques for applications including nighttime imaging and autonomous driving.

Technology Category

• Computer Vision: Computational Photography, Image & Video Synthesis
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Security and Privacy: Large-scale security measurements
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

We introduce the AIM 2025 Real-World RAW Image Denoising Challenge, aiming to advance efficient and effective denoising techniques grounded in data synthesis. The competition is built upon a newly established evaluation benchmark featuring challenging low-light noisy images captured in the wild using five different DSLR cameras. Participants are tasked with developing novel noise synthesis pipelines, network architectures, and training methodologies to achieve high performance across different camera models. Winners are determined based on a combination of performance metrics, including full-reference measures (PSNR, SSIM, LPIPS), and non-reference ones (ARNIQA, TOPIQ). By pushing the boundaries of camera-agnostic low-light RAW image denoising trained on synthetic data, the competition promotes the development of robust and practical models aligned with the rapid progress in digital photography. We expect the competition outcomes to influence multiple domains, from image restoration to night-time autonomous driving.

Problem

• Advancing efficient RAW image denoising via data synthesis
• Developing camera-agnostic models for low-light noisy images
• Creating robust denoising techniques for real-world photography applications

Innovation

• Develops novel noise synthesis pipelines for denoising
• Creates new network architectures for image restoration
• Introduces advanced training methodologies for cameras