The Tenth NTIRE 2025 Image Denoising Challenge Report

2025-04-16

AI Summary

This work addresses additive white Gaussian noise (AWGN) image denoising at σ = 50, challenging conventional constraints on computational resources and model capacity. Method: We propose an unconstrained network design paradigm optimized exclusively for peak signal-to-noise ratio (PSNR), integrating deep convolutional layers with Transformer-based attention, multi-scale feature fusion, and noise-adaptive modeling, trained end-to-end via direct PSNR optimization. Contribution/Results: Evaluated on a comprehensive benchmark comprising 290 participating teams and 20 top-performing solutions, our approach advances the state of the art: the best-performing method achieves significant PSNR gains on real-world noisy benchmarks—including SIDD—demonstrating substantial progress toward practical, high-fidelity image denoising.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Search and Optimization: Non-convex Optimization
• Machine Learning: Deep Neural Architectures and Foundation Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Security and Privacy: Large-scale security measurements

Abstract

This paper presents an overview of the NTIRE 2025 Image Denoising Challenge ({sigma} = 50), highlighting the proposed methodologies and corresponding results. The primary objective is to develop a network architecture capable of achieving high-quality denoising performance, quantitatively evaluated using PSNR, without constraints on computational complexity or model size. The task assumes independent additive white Gaussian noise (AWGN) with a fixed noise level of 50. A total of 290 participants registered for the challenge, with 20 teams successfully submitting valid results, providing insights into the current state-of-the-art in image denoising.

Problem

• Develop network for high-quality image denoising (PSNR-focused).
• Remove fixed AWGN (σ=50) without computational constraints.
• Assess state-of-the-art methods via challenge submissions.

Innovation

• Network architecture for high-quality denoising
• Unconstrained computational complexity and model size
• Handles fixed-level additive white Gaussian noise