Noise2Noise Revisited: Training Pair Distributions Dominate Loss Choice in Self-Supervised Denoising
cs.LG, cs.CV, eess.IV
Submitted: 2026-09-15
Updated: 2026-09-15
Comments: 8 pages, 3 figures, 3 tables. Accepted to The 8th International Conference on Video, Signal and Image Processing (VSIP 2026). Code and data: https://github.com/dyshang/noise2noise-revisited
Code: https://github.com/dyshang/noise2noise-revisited
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Next-Scale Prediction: A Self-Supervised Approach for Real-World Image Denoising
- Learning to Recorrupt: Noise Distribution Agnostic Self-Supervised Image Denoising
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: Local-Consistency Optimisation for Bayesian Neural Networks
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks