Learning Continuous Wasserstein Barycenter Space for Generalized All-in-One Image Restoration
IEEE Transactions on Pattern Analysis and Machine Intelligence · 2026-07 · 48 (7) · 8395-8410
DOI: 10.1109/tpami.2026.3669121 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Pattern Analysis and Machine Intelligence, 2026-07. Verfasst von Xiaole Tang und 4 weiteren Autor:innen. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 73 Literaturangaben. Gefördert von: National Key R&D Program, NSFC, National Postdoctoral Program for Innovative Talents.
Abstract (Originalsprache)
Despite substantial advances in all-in-one image restoration for addressing diverse degradations within a unified model, existing methods remain vulnerable to out-of-distribution degradations, thereby limiting their generalization in real-world scenarios. To tackle the challenge, this work is motivated by the intuition that multisource degraded feature distributions are induced by different degradation-specific shifts from an underlying degradation-agnostic distribution, and recovering such a shared distribution is thus crucial for achieving generalization across degradations. With this insight, we propose BaryIR, a representation learning framework that aligns multisource degraded features in the Wasserstein barycenter (WB) space, which models a degradation-agnostic distribution by minimizing the average of Wasserstein distances to multisource degraded distributions. We further introduce residual subspaces, whose embeddings are mutually contrasted while remaining orthogonal to the WB embeddings. Consequently, BaryIR explicitly decouples two orthogonal spaces: a WB space that encodes the degradation-agnostic invariant contents shared across degradations, and residual subspaces that adaptively preserve the degradation-specific knowledge. This disentanglement mitigates overfitting to in-distribution degradations and enables adaptive restoration grounded on the degradation-agnostic shared invariance. Extensive experiments demonstrate that BaryIR performs competitively against state-of-the-art all-in-one methods. Notably, BaryIR generalizes well to unseen degradations (e.g., types and levels) and shows remarkable robustness in learning generalized features, even when trained on limited degradation types and evaluated on real-world data with mixed degradations.
Förderung
- National Key R&D Program
- NSFC
- National Postdoctoral Program for Innovative Talents
- Fundamental Research Funds for the Central Universities, China
Zitieren
APA
Tang, X., He, X., Xu, J., Gu, X., & Sun, J. (2026). Learning Continuous Wasserstein Barycenter Space for Generalized All-in-One Image Restoration. IEEE Transactions on Pattern Analysis and Machine Intelligence, 48(7), 8395-8410. https://doi.org/10.1109/tpami.2026.3669121
Vancouver
Tang X, He X, Xu J, Gu X, Sun J. Learning Continuous Wasserstein Barycenter Space for Generalized All-in-One Image Restoration. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2026;48(7):8395-8410. doi:10.1109/tpami.2026.3669121
BibTeX
@article{Tang2026,
author = {Tang, Xiaole and He, Xiaoyi and Xu, Jiayi and Gu, Xiang and Sun, Jian},
title = {Learning Continuous Wasserstein Barycenter Space for Generalized All-in-One Image Restoration},
journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
year = {2026},
volume = {48},
number = {7},
pages = {8395--8410},
doi = {10.1109/tpami.2026.3669121}
}
EUVIMED