CoRe-UIE: Rethinking Coexisting and Region-wise Degradation for Underwater Image Enhancement

Chenghao Xu; Guanying Huo; Lin Chen; Weifeng Kong; Ziheng Cao

REVIEW 4 major objections 6 minor 51 references

CoRe-UIE: Rethinking Coexisting and Region-wise Degradation for Underwater Image Enhancement

T0 review · 4 major / 6 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper claims that underwater image enhancement should be modeled as region-wise collaboration among mechanism-specialized experts rather than a single global restoration mapping, and reports top scores on UIEB, LSUI, and U45.

desk verdict A competent MoE-for-underwater-enhancement paper whose central mechanism-awareness claim rests on undefined degradation cues P; the empirical results are real but the paper needs major revision before acceptance. read the letter →

arxiv 2608.08965 v1 pith:F53WZGZZ submitted 2026-08-10 cs.AI cs.CV

Weifeng Kong , Chenghao Xu , Lin Chen , Ziheng Cao , Guanying Huo This is my paper · ORCID

classification cs.AIcs.CV

keywords underwaterimageenhancementmixtureofexpertsregion-adaptiveroutingshared-backboneHSICrepresentationdisentanglementdegradationcuesrestoration

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Underwater photos are degraded in ways that vary from region to region: one area may have a green color cast, another residual haze, a third washed-out texture, and a fourth dark shadows, with several of these problems present in the same local patch. CoRe-UIE argues that a single network mapping applied uniformly to the whole image is the wrong tool for such compound, spatially mixed degradation. The paper proposes an expert-collaboration network: a shared expert preserves content everywhere, while four routable experts with identical architecture but independent weights specialize in color correction, scattering suppression, texture recovery, and illumination protection, selected per region by a Top-2 router. It reports the best scores among the compared methods on UIEB, LSUI, and U45, with UIEB PSNR of 26.88 dB, and supports the design with ablations that remove each expert, the shared expert, and the HSIC term. A sympathetic reading is that mechanism-aware region-wise routing, not a more powerful unified backbone, is the reason for the gains.

What carries the argument

The load-bearing mechanism is a region-adaptive Top-2 router over four shared-backbone routed experts, each a residual block with independent parameters, plus a shared expert active everywhere. The router combines learned features with four input-derived mechanism cues $\{P_c, P_{sc}, P_t, P_l\}$ for color imbalance, scattering contrast, texture structure, and illumination risk; a response alignment loss forces each expert's normalized spatial response to resemble its cue, and an HSIC-based loss penalizes statistical dependence among all routed-expert representation pairs. This is what carries the argument that same-architecture experts can specialize to different coexisting degradations without hand-designed heterogeneous structures.

What would settle it

Retrain CoRe-UIE on UIEB with the four mechanism cues replaced by random or constant per-pixel maps, keeping all other components fixed; if PSNR stays close to 26.88 dB, then cue-guided routing and response alignment are not carrying the claimed specialization, while a large drop would confirm that the missing cue formulas matter.

Watch

Extended reading notes

Core claim

CoRe-UIE claims that representing underwater restoration as region-adaptive collaboration among degradation-specific experts is both feasible and superior to uniform restoration. The framework keeps a content-preserving shared expert active for all spatial regions and four shared-backbone routed experts active only where their mechanism is relevant; a region-adaptive Top-2 router assigns normalized weights based on learned features plus input-derived mechanism cues. Two additional objectives make structurally identical experts diverge: a response alignment loss matches each expert's spatial response to its corresponding degradation cue, and an HSIC-based loss reduces statistical dependence among expert feature pairs. On UIEB the full model reports 26.88 dB PSNR, 0.9103 SSIM, 0.9666 FSIM, 0.9554 FSIMC, and 0.9855 VSI, and it reports the best compared scores on LSUI and U45 as well, so the paper's claim is that the routing-plus-disentanglement design, not a larger or differently shaped network, drives the improvement.

Load-bearing premise

The paper's routing and alignment losses are built entirely on four input-derived mechanism cues whose formulas are never given, so if those cues cannot be computed reliably from an input image, the mechanism-aware specialization has no definable foundation.

Editorial extensions

If this is right

On UIEB, the full model reaches 26.88 dB PSNR, 0.9103 SSIM, 0.9666 FSIM, 0.9554 FSIMC, and 0.9855 VSI, the best values among the compared methods.
On LSUI, CoRe-UIE also tops the compared methods on every full-reference metric, so the advantage is not tied to a single benchmark's degradation distribution.
On U45, which has no ground-truth references, CoRe-UIE gets the best NIQE, BRISQUE, CEIQ, and PIQE scores among the compared methods.
Ablations show each routed expert contributes; dropping the scattering expert causes the largest PSNR drop, and removing the HSIC term also degrades all reported metrics.
Top-2 routing outperforms Top-1, Top-3, and Top-4, supporting the claim that locally coexisting degradations need more than one expert but not dense activation of all experts.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

If the same recipe is transferred to other compound-degradation settings such as hazy low-light scenes, rain plus fog, or mixed noise and blur, the four mechanism cues would need to be redefined for each domain; the paper does not test this transfer.
Because the cue formulas are missing, the cleanest isolation experiment is one the paper does not run: compare input-derived cues against purely learned routing predictors with identical capacity and losses; that comparison would show how much of the gain is truly cue-guided.
A consequence the authors leave implicit is interpretability: response alignment should make each expert's spatial activation map readable as a where-is-this-degradation-happening signal, which could aid failure diagnosis in deployment, but the paper does not quantify this.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Dear colleague,

The paper is a solid, workmanlike application of mixture-of-experts to underwater image enhancement. What's new: it puts four shared-backbone experts behind a region-adaptive Top-k router, adds a shared content expert, and uses HSIC to decorrelate expert features. The benchmark results are competitive — best PSNR/SSIM/FSIM/FSIMC/VSI on UIEB and LSUI, best no-reference scores on U45 — and the ablations show each expert and the HSIC term earn their keep. The comparison is external, so there is no derivation-level circularity. The writing is clear and the motivation (spatially heterogeneous, coexisting degradations) is sensible.

The problem is the degradation cues P. Equations (4) and (9) depend on P for routing and for aligning expert responses, but P is never defined. The paper names four cues (color imbalance, scattering contrast, texture structure, illumination risk) and calls them input-derived, but gives no formulas, no estimation procedure, no normalization, no reference. Without P, the core claim — that experts specialize by degradation mechanism — cannot be implemented, reproduced, or even tested. This is load-bearing, not cosmetic. If P is learned from the features F, the 'physical cue' story collapses into circularity; if P is hand-crafted, the authors need to say exactly how. This is the one thing that keeps the paper from being acceptable.

There are also smaller but real issues. Table 1 lists Ours VSI=0.9855 on UIEB; Table 3 lists Ours(ToP-2) VSI=0.9821 for the same default model. That is a direct inconsistency. The reconstruction loss L_rec is described only as 'fidelity plus structural consistency,' not specified. k=2 is chosen by comparing Top-1/2/3/4 on the test sets, i.e., test-set selection. No error bars are reported. No code is released.

The gains over the runner-up are roughly 0.4 dB PSNR on UIEB, so the empirical impact is modest. Still, the architecture is a plausible new combination for a real subfield problem, and the ablations are informative. If the authors define P, fix the tables, and release code, this could be a solid contribution. As it stands, the central mechanism-aware routing claim is unsubstantiated.

My recommendation: send it to peer review, but require major revision. The referee should ask for a precise definition of P, corrected tables, and either code or a detailed experimental protocol. It's not a desk reject — the idea is worth examining — but it is not acceptable in this form.

Referee Report

4 major / 6 minor

Summary. CoRe-UIE proposes an underwater image enhancement framework based on a content-preserving shared expert and four same-architecture routed experts intended for color correction, scattering suppression, texture recovery, and illumination protection. A region-adaptive Top-k router assigns experts to spatial regions using input-derived mechanism cues, and training adds a response alignment loss plus an HSIC-based representation disentanglement term. Experiments on UIEB, LSUI, and U45 report state-of-the-art or competitive results on full-reference and no-reference metrics, alongside ablations of individual experts, the HSIC term, and the Top-k choice.

Significance. If the missing details are supplied, the paper would offer a plausible and potentially useful way to induce degradation-aware specialization in homogeneous mixture-of-experts architectures for low-level vision. The external benchmark comparisons and the carefully structured ablations are strengths, and the reported gains on UIEB and LSUI are substantial enough to be interesting. However, the central claim of mechanism-aware region routing currently rests on an undefined cue set, and the benchmark tables show an internal inconsistency, so the contribution is not yet fully verifiable.

major comments (4)

[Region-Adaptive Top-k Routing (Eqs. 4 and 8-9)] The four mechanism cues P={Pc, Psc, Pt, Pl} are never defined mathematically. The text only names them as color imbalance, scattering-related contrast degradation, texture structure, and illumination risk, and calls them 'input-derived,' but gives no formulas, estimation procedures, normalization, or references. Both the router G=Softmax(R(F,P)) in Eq. (4) and the response alignment loss in Eq. (9) depend on P, with Eq. (9) explicitly aligning Norm(R_i) to Norm(P_i). Without a precise definition of P, the method cannot be implemented, reproduced, or tested for whether the cues actually correspond to local degradations. Please provide exact definitions or clearly state that P is learned from F; if P is learned, the 'physical cue' interpretation must be revised.
[Loss Functions (Eq. 7)] The reconstruction loss L_rec is described only as combining 'pixel-wise fidelity terms with structural consistency constraints' and is never written out. Since the quantitative benchmark results are the main evidence for the method's effectiveness, the exact form of L_rec (e.g., L1, L2, SSIM, perceptual terms, and their weights) is needed for reproducibility and for assessing whether the improvement comes from the proposed routing or from an ad hoc reconstruction objective. Please specify L_rec explicitly and report the values of lambda_align and lambda_hsic.
[Tables 1 and 3] The default model's VSI on UIEB is reported as 0.9855 in Table 1 but as 0.9821 in Table 3 for 'Ours(ToP-2)', which is the same default configuration. These values cannot both be correct. Moreover, no error bars or significance tests are reported anywhere, so it is unclear whether the differences from the second-best methods are stable across runs. Please reconcile the inconsistent VSI values and provide variance information or repeated-run statistics.
[Ablation Study (Table 3)] The ablations remove whole experts or the HSIC term but never vary or ablate the mechanism cues P. Consequently, the experiments do not demonstrate that routing decisions correspond to the intended degradation mechanisms; they only show that having more experts helps. To support the central claim of mechanism-aware region-adaptive routing, please include analyses that visualize or quantify routing maps, cue maps, or per-expert response heatmaps, and, if possible, a variant in which P is replaced by randomized or ablated cues.

minor comments (6)

[Author line, page 1] The author line 'Ziheng Cao1 Guanying Huo1∗' is missing a comma between the names 'Ziheng Cao1' and 'Guanying Huo1'.
[Table 3 and text] The label 'Ours(ToP-2)' in Table 3 should be 'Ours (Top-2)' to match the terminology used in the text.
[Related Work] The Related Work section states that CoRe-UIE introduces 'four structurally differentiated routed experts,' but the Methodology consistently says all routed experts share the same architecture with independent parameters; please align this wording.
[Tables 1-3] Method names are inconsistent across tables: 'SSUIE' appears in the tables while 'SS-UIE' is used in the text; please unify the naming.
[Equation (3)] The expert block in Eq. (3) does not specify the nonlinearity sigma or the kernel sizes and normalization for the convolutional layers C1_i and C2_i; please provide these implementation details in the setup section.
[Equation (9)] The spatial normalization Norm(·) used in the response alignment loss is not defined; please specify whether it is per-channel or per-region and how it is computed in practice.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: benchmark results rest on external comparisons, and expert specialization is an explicitly enforced training objective rather than a derived prediction.

full rationale

CoRe-UIE's main quantitative claims are established by comparing against external baselines on UIEB, LSUI, and U45 (Tables 1-2); these results are not derived from the model's own assumptions and are therefore self-contained evidence. The mechanism-expert specialization is not presented as an emergent prediction: Eq. (9) explicitly trains each expert response to align with the input-derived cues P, so any correspondence between expert roles and cue names is an enforced training objective, not a logically circular reduction. The HSIC loss and Top-k routing are architectural/training choices whose effects are evaluated by ablations (Table 3). The undefinedness of P in Eqs. (4), (8), and (9) is a reproducibility and completeness gap, but no equation equates a predicted quantity to a fitted input by construction, and no load-bearing self-citation or imported uniqueness theorem is used. The paper is therefore not circular, though the mechanism-cue computation would need to be specified for full reproducibility.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The framework's core ideas (region-wise routing, cue-guided alignment, HSIC decorrelation) rest on unstated cue estimation and unreported hyperparameters. No new physical entities are introduced.

free parameters (2)

Top-k value k = 2 (selected via ablation on test benchmarks)
The paper compares Top-1, Top-2, Top-3, and Top-4 in Table 3 and selects k=2 as default; this is a hyperparameter tuned on the evaluation set.
Loss weights lambda_align and lambda_hsic = Not reported
Equation (7) defines the joint objective with two regularization weights, but the paper never gives their values, so the exact training objective is underdetermined.

assumptions (4)

domain assumption Input-derived mechanism cues accurately encode local degradation types
Used in Eqs. 4-9 to route experts and align responses; no definition or validation is provided.
domain assumption Reducing HSIC among expert representations improves restoration quality
HSIC loss Eq. 12 is added to decorrelate experts; only one ablation (w/o HSIC) supports it, with no theoretical guarantee.
domain assumption Paired reference images are reliable supervision targets
Training and full-reference evaluation assume UIEB and LSUI references are correct enhanced images.
standard math L_rec is a standard pixel plus structural loss
Equation (7) references common practice and Zamir et al. 2022, but the exact composition of L_rec is omitted.

how reviews work

0 comments

Cite this review

Pith. "Pith review of CoRe-UIE: Rethinking Coexisting and Region-wise Degradation for Underwater Image Enhancement." pith.science (2026). https://pith.science/paper/F53WZGZZ

@misc{pith2026260808965,
  author       = {Pith},
  title        = {Pith review of: CoRe-UIE: Rethinking Coexisting and Region-wise Degradation for Underwater Image Enhancement},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/F53WZGZZ}},
  note         = {Machine review of arXiv:2608.08965}
}

abstract

Underwater images often suffer from diverse and coexisting degradations, including color distortion, scattering haze, texture attenuation, and uneven illumination. These degradations vary across regions and may coexist locally, making conventional uniform restoration difficult to adapt to different degradation patterns. To address this problem, we propose Coexisting and Region-wise Degradation for Underwater Image Enhancement (\textbf{CoRe-UIE}), a degradation-oriented expert collaboration framework. CoRe-UIE combines a content-preserving shared expert with four shared-backbone routed experts for color correction, scattering suppression, texture recovery, and illumination protection. The routed experts share the same architecture but have independent parameters, and are assigned to different regions through input-derived degradation cues and region-adaptive Top-$k$ routing. We further introduce a Hilbert--Schmidt Independence Criterion (HSIC)-based representation constraint to reduce statistical dependence among expert features and alleviate redundant expert responses. Experiments on UIEB, LSUI, and U45 demonstrate that CoRe-UIE achieves competitive quantitative performance and visually balanced enhancement under diverse underwater degradation conditions.

Figures

Figures reproduced from arXiv: 2608.08965 by the authors.

**Figure 1.** Motivation of CoRe-UIE. adaptive collaboration among different restoration mechanisms rather than a one-size-fits-all mapping. Previous underwater enhancement methods have explored physical imaging models, handcrafted priors, and multi-scale fusion to compensate for wavelength-dependent attenuation and scattering (Akkaynak and Treibitz 2019; Li et al. 2020). However, their performance is often sensitive to assumpt… view at source ↗

**Figure 2.** Overall architecture of the proposed CoRe-UIE framework. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

**Figure 3.** Qualitative comparison of underwater image enhancement results. [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗

Figures from the paper (2 more)

**Figure 5.** Figure 5: Qualitative comparison of different Top- [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]

**Figure 4.** Figure 4: Qualitative ablation study of mechanism-guided [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

51 extracted references · 46 canonical work pages

[1]

An Underwater Image Enhancement Benchmark Dataset and Beyond

Li, Chongyi and Guo, Chunle and Ren, Wenqi and Cong, Runmin and Hou, Junhui and Kwong, Sam and Tao, Dacheng. An Underwater Image Enhancement Benchmark Dataset and Beyond. IEEE Transactions on Image Processing

work page
[2]

U-Shape Transformer for Underwater Image Enhancement

Peng, Lintao and Zhu, Chunli and Bian, Liheng. U-Shape Transformer for Underwater Image Enhancement. IEEE Transactions on Image Processing

work page
[3]

Sea-Thru: A Method for Removing Water From Underwater Images

Akkaynak, Derya and Treibitz, Tali. Sea-Thru: A Method for Removing Water From Underwater Images. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

work page
[4]

Underwater Image Enhancement via Medium Transmission-Guided Multi-Color Space Embedding

Li, Chongyi and Anwar, Saeed and Hou, Junhui and Cong, Runmin and Guo, Chunle and Ren, Wenqi. Underwater Image Enhancement via Medium Transmission-Guided Multi-Color Space Embedding. IEEE Transactions on Image Processing

work page
[5]

Uncertainty Inspired Underwater Image Enhancement

Fu, Zhenqi and Wang, Wu and Huang, Yue and Ding, Xinghao and Ma, Kai-Kuang. Uncertainty Inspired Underwater Image Enhancement. European Conference on Computer Vision ECCV 2022. 2022

work page 2022
[6]

Contrastive Semi-Supervised Learning for Underwater Image Restoration via Reliable Bank

Huang, Shirui and Wang, Keyan and Liu, Huan and Chen, Jun and Li, Yunsong. Contrastive Semi-Supervised Learning for Underwater Image Restoration via Reliable Bank. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

work page
[7]

Five A ^ + Network: You Only Need 9K Parameters for Underwater Image Enhancement

Jiang, Jingxia and Ye, Tian and Bai, Jinbin and Chen, Sixiang and Chai, Wenhao and Jun, Shi and Liu, Yun and Chen, Erkang. Five A ^ + Network: You Only Need 9K Parameters for Underwater Image Enhancement. British Machine Vision Conference (BMVC)

work page
[8]

and Vipparthi, Santosh Kumar and Nandi, Sukumar

Khan, Md Raqib and Mishra, Priyanka and Mehta, Nancy and Phutke, Shruti S. and Vipparthi, Santosh Kumar and Nandi, Sukumar. Spectroformer: Multi-Domain Query Cascaded Transformer Network for Underwater Image Enhancement. Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)

work page

Show all 51 references

[9]

Wavelet-Based Fourier Information Interaction with Frequency Diffusion Adjustment for Underwater Image Restoration

Zhao, Chen and Cai, Weiling and Dong, Chenyu and Hu, Chengwei. Wavelet-Based Fourier Information Interaction with Frequency Diffusion Adjustment for Underwater Image Restoration. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
[10]

WaterMamba: Visual State Space Model for Underwater Image Enhancement

Guan, Meisheng and Xu, Haiyong and Jiang, Gangyi and Yu, Mei and Chen, Yeyao and Luo, Ting and Song, Yang. WaterMamba: Visual State Space Model for Underwater Image Enhancement. arXiv preprint arXiv:2405.08419

arXiv
[11]

O-Mamba: O-Shape State-Space Model for Underwater Image Enhancement

Dong, Chenyu and Zhao, Chen and Cai, Weiling and Yang, Bo. O-Mamba: O-Shape State-Space Model for Underwater Image Enhancement. arXiv preprint arXiv:2408.12816

arXiv
[12]

Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer

Shazeer, Noam and Mirhoseini, Azalia and Maziarz, Krzysztof and Davis, Andy and Le, Quoc and Hinton, Geoffrey and Dean, Jeff. Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer. International Conference on Learning Representations (ICLR)
[13]

GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding

Lepikhin, Dmitry and Lee, HyoukJoong and Xu, Yuanzhong and Chen, Dehao and Firat, Orhan and Huang, Yanping and Krikun, Maxim and Shazeer, Noam and Chen, Zhifeng. GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding. International Conference on Learn...
[14]

Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity

Fedus, William and Zoph, Barret and Shazeer, Noam. Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity. Journal of Machine Learning Research
[15]

Scaling Vision with Sparse Mixture of Experts

Riquelme, Carlos and Puigcerver, Joan and Mustafa, Basil and Neumann, Maxim and Jenatton, Rodolphe and Pinto, Andr \'e Susano and Keysers, Daniel and Houlsby, Neil. Scaling Vision with Sparse Mixture of Experts. Advances in Neural Information Processing Systems (NeurIPS)
[16]

From Sparse to Soft Mixtures of Experts

Puigcerver, Joan and Riquelme, Carlos and Mustafa, Basil and Houlsby, Neil. From Sparse to Soft Mixtures of Experts. International Conference on Learning Representations (ICLR)
[17]

Complexity Experts are Task-Discriminative Learners for Any Image Restoration

Zamfir, Eduard and Wu, Zongwei and Mehta, Nancy and Tan, Yuedong and Paudel, Danda Pani and Zhang, Yulun and Timofte, Radu. Complexity Experts are Task-Discriminative Learners for Any Image Restoration. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Reco...
[18]

Measuring Statistical Dependence with Hilbert-Schmidt Norms

Gretton, Arthur and Bousquet, Olivier and Smola, Alex and Sch \"o lkopf, Bernhard. Measuring Statistical Dependence with Hilbert-Schmidt Norms. International Conference on Algorithmic Learning Theory (ALT)
[19]

Iterative Predictor-Critic Code Decoding for Real-World Image Dehazing

Fu, Jiayi and Liu, Siyu and Liu, Zikun and Guo, Chun-Le and Park, Hyunhee and Wu, Ruiqi and Wang, Guoqing and Li, Chongyi. Iterative Predictor-Critic Code Decoding for Real-World Image Dehazing. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
[20]

Color Correction Meets Cross-Spectral Refinement: A Distribution-Aware Diffusion for Underwater Image Restoration

Chang, Laibin and Wang, Yunke and Du, Bo and Xu, Chang. Color Correction Meets Cross-Spectral Refinement: A Distribution-Aware Diffusion for Underwater Image Restoration. IEEE Transactions on Multimedia
[21]

Underwater Image Enhancement With Cascaded Contrastive Learning

Liu, Yi and Jiang, Qiuping and Wang, Xinyi and Luo, Ting and Zhou, Jingchun. Underwater Image Enhancement With Cascaded Contrastive Learning. IEEE Transactions on Multimedia
[22]

See Through Water: Heuristic Modeling Toward Color Correction for Underwater Image Enhancement

Fan, Junyu and Xu, Jie and Zhou, Jingchun and Meng, Danling and Lin, Yi. See Through Water: Heuristic Modeling Toward Color Correction for Underwater Image Enhancement. IEEE Transactions on Circuits and Systems for Video Technology
[23]

Reti-Diff: Illumination Degradation Image Restoration with Retinex-Based Latent Diffusion Model

He, Chunming and Fang, Chengyu and Zhang, Yulun and Tang, Longxiang and Huang, Jinfa and Li, Kai and Guo, zhenhua and Li, Xiu and Farsiu, Sina. Reti-Diff: Illumination Degradation Image Restoration with Retinex-Based Latent Diffusion Model. International Conference on Learning...
[24]

You Only Look Yourself: Unsupervised and Untrained Single Image Dehazing Neural Network

Li, Boyun and Gou, Yuanbiao and Gu, Shuhang and Liu, Jerry Zitao and Zhou, Joey Tianyi and Peng, Xi. You Only Look Yourself: Unsupervised and Untrained Single Image Dehazing Neural Network. International Journal of Computer Vision
[25]

Rank-One Prior: Toward Real-Time Scene Recovery

Liu, Jun and Liu, Ryan Wen and Sun, Jianing and Zeng, Tieyong. Rank-One Prior: Toward Real-Time Scene Recovery. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
[26]

Rank-One Prior: Real-Time Scene Recovery

Liu, Jun and Liu, Ryan Wen and Sun, Jianing and Zeng, Tieyong. Rank-One Prior: Real-Time Scene Recovery. IEEE Transactions on Pattern Analysis and Machine Intelligence
[27]

ALSP+: Fast Scene Recovery via Ambient Light Similarity Prior

He, Lei and Yi, Zunhui and Liu, Jinshi and Chen, Chaoyang and Lu, Ming and Chen, Zhipeng. ALSP+: Fast Scene Recovery via Ambient Light Similarity Prior. IEEE Transactions on Image Processing
[28]

and Sheikh, Hamid R

Wang, Zhou and Bovik, Alan C. and Sheikh, Hamid R. and Simoncelli, Eero P. Image Quality Assessment: From Error Visibility to Structural Similarity. IEEE Transactions on Image Processing
[29]

FSIM: A Feature Similarity Index for Image Quality Assessment

Zhang, Lin and Zhang, Lei and Mou, Xuanqin and Zhang, David. FSIM: A Feature Similarity Index for Image Quality Assessment. IEEE Transactions on Image Processing
[30]

VSI: A Visual Saliency-Induced Index for Perceptual Image Quality Assessment

Zhang, Lin and Shen, Ying and Li, Hongyu. VSI: A Visual Saliency-Induced Index for Perceptual Image Quality Assessment. IEEE Transactions on Image Processing
[31]

An Underwater Color Image Quality Evaluation Metric

Yang, Miao and Sowmya, Arcot. An Underwater Color Image Quality Evaluation Metric. IEEE Transactions on Image Processing
[32]

Human-Visual-System-Inspired Underwater Image Quality Measures

Panetta, Karen and Gao, Chen and Agaian, Sos. Human-Visual-System-Inspired Underwater Image Quality Measures. IEEE Journal of Oceanic Engineering
[33]

An Imaging-Inspired No-Reference Underwater Color Image Quality Assessment Metric

Wang, Yan and Li, Na and Li, Zongying and Gu, zhaorui and Zheng, Haiyong and Zheng, Bing and Sun, Mengmam. An Imaging-Inspired No-Reference Underwater Color Image Quality Assessment Metric. Computers & Electrical Engineering
[34]

Restormer: Efficient Transformer for High-Resolution Image Restoration

Zamir, Syed Waqas and Arora, Aditya and Khan, Salman and Hayat, Munawar and Khan, Fahad Shahbaz and Yang, Ming-Hsuan. Restormer: Efficient Transformer for High-Resolution Image Restoration. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
[35]

Completely Blind

Mittal, Anish and Soundararajan, Rajiv and Bovik, Alan C. Making a “Completely Blind” Image Quality Analyzer. IEEE Signal Processing Letters. 2013. doi:10.1109/LSP.2012.2227726

2013
[36]

No-Reference Image Quality Assessment in the Spatial Domain

Mittal, Anish and Moorthy, Anush Krishna and Bovik, Alan Conrad. No-Reference Image Quality Assessment in the Spatial Domain. IEEE Transactions on Image Processing. 2012. doi:10.1109/TIP.2012.2214050

2012
[37]

No-Reference Quality Assessment of Contrast-Distorted Images Using Contrast Enhancement

Yan, Jia and Li, Jie and Fu, Xin. No-Reference Quality Assessment of Contrast-Distorted Images Using Contrast Enhancement. 2019. arXiv:1904.08879

2019 arXiv
[38]

and Praneeth, D

Venkatanath, N. and Praneeth, D. and Bh, Maruthi Chandrasekhar and Channappayya, Sumohana S. and Medasani, S. S. Blind Image Quality Evaluation Using Perception Based Features. Proceedings of the Twenty First National Conference on Communications (NCC). 2015. doi:10.1109/NCC.2...

2015
[39]

2019 , eprint =

A Fusion Adversarial Underwater Image Enhancement Network with a Public Test Dataset , author =. 2019 , eprint =

2019
[40]

IEEE Transactions on Image Processing , volume =

Task-Driven Underwater Image Enhancement via Hierarchical Semantic Refinement , author =. IEEE Transactions on Image Processing , volume =. 2026 , doi =

2026
[41]

IEEE Transactions on Geoscience and Remote Sensing , pages =

UEAOD: Underwater Image Enhancement-Assisted Object Detection Guided by Frequency Decomposition and Physical Consistency , author =. IEEE Transactions on Geoscience and Remote Sensing , pages =. 2026 , doi =

2026
[42]

ACM Computing Surveys , volume =

Zhu, Rongxin and Sheng, Lei and Wu, Kaitao and Boukerche, Azzedine and Long, Libo and Yang, Qiuling , title =. ACM Computing Surveys , volume =. 2026 , doi =

2026
[43]

IEEE Transactions on Image Processing , volume =

Xue, Xinwei and Xu, Zimeng and Wang, Kaixin and Yuan, Jincheng and Zhou, Jingchun and Xu, Chengpei and Shang, Xiaoke and Ma, Long and Wang, Weimin , title =. IEEE Transactions on Image Processing , volume =. 2026 , doi =

2026
[44]

IEEE Access , volume =

Yu, Hansung and Vo, Chuong Hoang and Lee, Chul , title =. IEEE Access , volume =. 2026 , doi =

2026
[45]

, title =

Wen, Junjie and Yang, Guidong and Zhao, Benyun and Lei, Lei and Gao, Zhi and Chen, Xi and Chen, Ben M. , title =. IEEE Journal of Oceanic Engineering , volume =. 2026 , doi =

2026
[46]

Journal of Water Process Engineering , volume =

Si, Jongwook and Kim, Sungyoung , title =. Journal of Water Process Engineering , volume =. 2026 , doi =

2026
[47]

Engineering Applications of Artificial Intelligence , volume =

Zhai, Lujun and Zhu, Ye and Russakovsky, Olga and Cui, Suxia , title =. Engineering Applications of Artificial Intelligence , volume =. 2026 , doi =

2026
[48]

Proceedings of the AAAI Conference on Artificial Intelligence , volume =

Tan, Zijun and Fu, Chuan and Guo, Tan and Nan, Zhixiong and Zhou, Pengzhan and Peng, Xinggan and Luo, Fulin , title =. Proceedings of the AAAI Conference on Artificial Intelligence , volume =. 2026 , doi =

2026
[49]

Proceedings of the AAAI Conference on Artificial Intelligence , volume =

Peng, Lintao and Bian, Liheng , title =. Proceedings of the AAAI Conference on Artificial Intelligence , volume =. 2025 , doi =

2025
[50]

A Comprehensive Survey on Underwater Image Enhancement Based on Deep Learning , year=

Cong, Xiaofeng and Zhao, Yu and Gui, Jie and Hou, Junming and Tao, Dacheng , journal=. A Comprehensive Survey on Underwater Image Enhancement Based on Deep Learning , year=
[51]

2020 , author =

Diving deeper into underwater image enhancement: A survey , journal =. 2020 , author =

2020

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [1]

An Underwater Image Enhancement Benchmark Dataset and Beyond

Li, Chongyi and Guo, Chunle and Ren, Wenqi and Cong, Runmin and Hou, Junhui and Kwong, Sam and Tao, Dacheng. An Underwater Image Enhancement Benchmark Dataset and Beyond. IEEE Transactions on Image Processing

work page

[2] [2]

U-Shape Transformer for Underwater Image Enhancement

Peng, Lintao and Zhu, Chunli and Bian, Liheng. U-Shape Transformer for Underwater Image Enhancement. IEEE Transactions on Image Processing

work page

[3] [3]

Sea-Thru: A Method for Removing Water From Underwater Images

Akkaynak, Derya and Treibitz, Tali. Sea-Thru: A Method for Removing Water From Underwater Images. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

work page

[4] [4]

Underwater Image Enhancement via Medium Transmission-Guided Multi-Color Space Embedding

Li, Chongyi and Anwar, Saeed and Hou, Junhui and Cong, Runmin and Guo, Chunle and Ren, Wenqi. Underwater Image Enhancement via Medium Transmission-Guided Multi-Color Space Embedding. IEEE Transactions on Image Processing

work page

[5] [5]

Uncertainty Inspired Underwater Image Enhancement

Fu, Zhenqi and Wang, Wu and Huang, Yue and Ding, Xinghao and Ma, Kai-Kuang. Uncertainty Inspired Underwater Image Enhancement. European Conference on Computer Vision ECCV 2022. 2022

work page 2022

[6] [6]

Contrastive Semi-Supervised Learning for Underwater Image Restoration via Reliable Bank

Huang, Shirui and Wang, Keyan and Liu, Huan and Chen, Jun and Li, Yunsong. Contrastive Semi-Supervised Learning for Underwater Image Restoration via Reliable Bank. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

work page

[7] [7]

Five A ^ + Network: You Only Need 9K Parameters for Underwater Image Enhancement

Jiang, Jingxia and Ye, Tian and Bai, Jinbin and Chen, Sixiang and Chai, Wenhao and Jun, Shi and Liu, Yun and Chen, Erkang. Five A ^ + Network: You Only Need 9K Parameters for Underwater Image Enhancement. British Machine Vision Conference (BMVC)

work page

[8] [8]

and Vipparthi, Santosh Kumar and Nandi, Sukumar

Khan, Md Raqib and Mishra, Priyanka and Mehta, Nancy and Phutke, Shruti S. and Vipparthi, Santosh Kumar and Nandi, Sukumar. Spectroformer: Multi-Domain Query Cascaded Transformer Network for Underwater Image Enhancement. Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)

work page