UNet-AF: An alias-free UNet for image restoration

2026-03-11

AI Summary

This work addresses the limited translation equivariance of conventional UNet architectures, which stems from aliasing-prone convolutions and downsampling operations, thereby constraining their performance in image restoration tasks. The paper presents the first systematically designed fully alias-free UNet by replacing standard non-equivariant components with advanced translation-equivariant convolution and up/downsampling layers, ensuring strict equivariance throughout the entire network. Experimental results demonstrate that the proposed architecture achieves restoration performance on par with non-equivariant baselines while significantly improving empirical translation equivariance. Ablation studies further confirm the necessity of each equivariant component in realizing these gains.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Deep Neural Architectures and Foundation Models
• Search and Optimization: Non-convex Optimization

Application Category

• Responsible Web: Algorithmic accountability and transparency on the web
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems

Abstract

The simplicity and effectiveness of the UNet architecture makes it ubiquitous in image restoration, image segmentation, and diffusion models. They are often assumed to be equivariant to translations, yet they traditionally consist of layers that are known to be prone to aliasing, which hinders their equivariance in practice. To overcome this limitation, we propose a new alias-free UNet designed from a careful selection of state-of-the-art translation-equivariant layers. We evaluate the proposed equivariant architecture against non-equivariant baselines on image restoration tasks and observe competitive performance with a significant increase in measured equivariance. Through extensive ablation studies, we also demonstrate that each change is crucial for its empirical equivariance. Our implementation is available at https://github.com/jscanvic/UNet-AF

Problem

• UNet
• aliasing
• translation equivariance
• image restoration

Innovation

• alias-free
• translation-equivariant
• UNet
• image restoration
• equivariance