UniH$^3$: Unifying Hierarchical Homogeneity and Heterogeneity for All-in-One Medical Image Restoration

Abstract

All-in-One medical image restoration (MedIR) aims to address diverse tasks across modalities and degradation types using a single universal model. Existing methods typically prioritize modeling inter-task heterogeneity (e.g., distinct data distributions and degradation types). However, they largely neglect the inherent homogeneity present in medical images, such as widely shared anatomical structures within and across modalities, which can be leveraged to ease model training and improve generalization. To this end, we propose UniH3, a novel framework that Unifies Hierarchical Homogeneity and Heterogeneity for all-in-one medical image restoration. Specifically, to comprehensively exploit homogeneity, we introduce a Hierarchical Homogeneity Memory (H2M) module that progressively distills intra- and inter-task homogeneity priors from high-quality images during training, and adaptively retrieves the most relevant priors tailored to the input for guided restoration. These retrieved priors are then injected into the restoration pipeline via an efficient Homogeneity-Guided Attention (HGA) mechanism. Furthermore, to comprehensively address heterogeneity, we design a Hierarchical Heterogeneity Balancer (H2B) that mitigates both inter- and intra-task conflicts during optimization, facilitating balanced and effective multi-task learning. Extensive experiments on two large-scale benchmarks, MedIR-2D-500K and MedIR-3D-3K, demonstrate that UniH3 achieves state-of-the-art performance on both all-in-one and single-task medical image restoration. We hope this work establishes a strong benchmark and advances the development of general-purpose medical image restoration models. Code is available at https://github.com/Yaziwel/UniH3.

Authors 6

Zhiwen Yang, Jiayin Li, Chengyu Liu, Hui Zhang, Bingzheng Wei, Yan Xu

Specification

Official page

arxiv.org/abs/2609.11156

Source:arXiv (Atom API + RSS)·T1·observed 3 h ago·high

Arxiv announce type

new

Source:arXiv (Atom API + RSS)·T1·observed 3 h ago·high

arXiv id

2609.11156

Source:arXiv (Atom API + RSS)·T1·observed 3 h ago·high

Categories

cs.CV

Source:arXiv (Atom API + RSS)·T1·observed 3 h ago·high

Github repo

Yaziwel/UniH3

Source:Hugging Face Hub (public pages, model cards, papers)·T2·observed 3 h ago·medium

Hf paper url

https://huggingface.co/papers/2609.11156

Source:Hugging Face Hub (public pages, model cards, papers)·T2·observed 3 h ago·medium

Github stars

5

Source:Hugging Face Hub (public pages, model cards, papers)·T2·observed 3 h ago·medium

Hf comments

1

Source:Hugging Face Hub (public pages, model cards, papers)·T2·observed 3 h ago·medium

Upvotes

2

Source:Hugging Face Hub (public pages, model cards, papers)·T2·observed 3 h ago·medium

PDF

arxiv.org/pdf/2609.11156

Source:arXiv (Atom API + RSS)·T1·observed 3 h ago·high

Primary category

cs.CV

Source:arXiv (Atom API + RSS)·T1·observed 3 h ago·high

Published

11 Sept 2026

Source:arXiv (Atom API + RSS)·T1·observed 3 h ago·high

Each value shows its source, tier and observation time. Conflicting claims are kept side by side and flagged — never averaged. How AI Atlas records facts →

Provenance