W2WDiff: Generalizing Underwater Diffusion Model via Unsupervised Underwater Conversion
IEEE Transactions on Geoscience and Remote Sensing · 2026 · 64 · 1-13
DOI: 10.1109/tgrs.2025.3629979 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Geoscience and Remote Sensing, 2026. Verfasst von Yuanlin Zhang und 6 weiteren Autor:innen. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 46 Literaturangaben. Gefördert von: National Key Research and Development Program of China, Project of Minzu University of China, Natural Science Foundation of Tianjin, China.
Abstract (Originalsprache)
Underwater image enhancement (UIE) is essential for underwater information acquisition across marine science and ocean remote sensing. Diffusion-based UIE methods demonstrate remarkable enhancement capabilities but suffer significant performance degradation when test-time observations diverge from training-time assumptions. Moreover, the inherent stochasticity of the diffusion process often manifests as inconsistent and unstable enhancement results, compromising both reproducibility and quality assurance. To address these limitations, we propose water-to-water diffusion (W2WDiff), a novel framework that introduces an unsupervised water-to-water (W2W) transformation strategy. By mapping heterogeneous underwater degradations to a tractable intermediate domain, our method circumvents the distribution shift problem inherent in direct enhancement approaches, achieving superior generalization across diverse underwater scenarios. In contrast to general pixel-space methods, we establish the feasibility of latent-space UIE and introduce a corresponding diffusion paradigm. Our approach introduces a custom Markov chain specifically designed for underwater characteristics, achieving substantial reductions in sampling steps while mitigating color distortion. Furthermore, we propose a three-stage training scheme along with a content consistency module (CCM) to mitigate pixel-level misalignment and enhance local structural fidelity and detail preservation. Comprehensive experiments demonstrate that W2WDiff achieves consistent and robust enhancement across a wide range of challenging underwater conditions, exhibiting strong zero-shot generalization performance. Project Page: https://c-pupil.github.io/projects/W2WDiff/index.html
Förderung
- National Key Research and Development Program of China
- Project of Minzu University of China
- Natural Science Foundation of Tianjin, China
- Fundamental Research Funds for the Central Universities
- National Natural Science Foundation of China
- China Postdoctoral Science Foundation—Tianjin Joint Support Program
- Tianjin Natural Science Foundation Project
Zitieren
APA
Zhang, Y., Yuan, J., Chen, X., Tang, X., Chen, Q., Wang, Y., & Li, C. (2026). W2WDiff: Generalizing Underwater Diffusion Model via Unsupervised Underwater Conversion. IEEE Transactions on Geoscience and Remote Sensing, 64, 1-13. https://doi.org/10.1109/tgrs.2025.3629979
Vancouver
Zhang Y, Yuan J, Chen X, Tang X, Chen Q, Wang Y, et al. W2WDiff: Generalizing Underwater Diffusion Model via Unsupervised Underwater Conversion. IEEE Transactions on Geoscience and Remote Sensing. 2026;64:1-13. doi:10.1109/tgrs.2025.3629979
BibTeX
@article{Zhang2026,
author = {Zhang, Yuanlin and Yuan, Jieyu and Chen, Xiao and Tang, Xiongxin and Chen, Qiao and Wang, Yiquan and Li, Chongyi},
title = {W2WDiff: Generalizing Underwater Diffusion Model via Unsupervised Underwater Conversion},
journal = {IEEE Transactions on Geoscience and Remote Sensing},
year = {2026},
volume = {64},
pages = {1--13},
doi = {10.1109/tgrs.2025.3629979}
}
EUVIMED