- Rating
- 4
- Confidence
- 4
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 3
Read review
Paper summary
This article proposes UG Prior, an uncertainty guided framework for zero sample low light image enhancement. The author points out that the existing illumination invariant priors may have unreliable estimates in severely degraded low light environments, making it difficult to balance robustness and information richness. To address this issue, the author proposes uncertainty guided multi-scale prior fusion in this article and introduces a color reliability weighting mechanism to preserve reliable structural and color information. The author designed a prior decoding regularization and online prior image collaborative optimization mechanism to further improve the prior utilization and dynamic updating in the generation process. The experiment was conducted on multiple low light image enhancement datasets, and the results showed that the method has good performance in image restoration quality and cross dataset generalization.
Critical strengths
The research method design is targeted. This article focuses on the reliability of illumination invariant priors under severe low illumination degradation conditions, and balances the robustness and information richness of priors through uncertainty guided multi-scale fusion. Compared to directly using a fixed lighting invariant prior, this adaptive construction method has a reasonable motivation and also reflects a certain degree of innovation.
The method design is complete and the experimental results are convincing. This article not only improves prior construction, but also enhances prior utilization in the generation process through prior decoding regularization and online update mechanisms. The author conducted quantitative and qualitative comparisons on multiple datasets and provided ablation experiments, visualization analysis, and model distillation experiments. The overall results support the effectiveness and certain cross dataset generalization ability of the proposed method.
Critical weaknesses
The effectiveness of uncertainty estimation still lacks direct verification. This article constructs uncertainty maps through brightness and local structure statistics, and uses them to guide multi-scale prior fusion. Although ablation experiments have shown that this design can improve the final enhancement performance, there is still a lack of direct evidence to prove a good correspondence between the estimated uncertainty and the true prior error. This to some extent weakens the explanation of the core mechanism.
The effectiveness of the online prior image collaborative optimization mechanism still requires more direct evidence. This article predicts priors through an auxiliary decoder and performs weighted updates during the generation process. Although ablation experiments have shown that online updates can improve the final results, it has not been fully demonstrated whether prior representations gradually become more accurate with iteration. In addition, this update mechanism is mainly based on weighted fusion, and its difference from general iterative prior correction methods can be further clarified.
Questions for authors
-
Section 3.2 estimates uncertainty using brightness and structural information. Has the correlation performance between the uncertainty graph and the true prior error been verified?
-
Section 3.4 achieves online prior updates through weighted fusion. Can the prior error changes under different generation steps be displayed to verify whether the updated prior is indeed more accurate?
-
When using QuadPrior prior in Table 3, the PSNR decreased from 21.29 dB to 19.10 dB. Did the experiment maintain the same generation backbone and training settings to ensure that the performance difference mainly came from the prior itself?
-
Appendix D.2 mentions adjusting the output of other methods uniformly to 1024 resolution. Will this scaling affect the evaluation results when using the same test images and partitions for all methods?
-
Defogging and underwater image enhancement can use the same theory as low light image enhancement. Can your paper verify the effectiveness of these two tasks. Note: SOTA is not required here, only the universality of your framework needs to be verified.
I suggest accepting this article clearly. But you need to discuss the above issues in detail.
6 comments
Official Response to Reviewer mrSq (Part 2/3)
Q4: Does resizing affect the evaluation?
We apologize for the incorrect description of the evaluation resolution. All comparison methods use their official inference commands, and outputs are resized to the original GT dimensions for evaluation. UG-Prior generates images at 1024 × 1024, followed by bicubic resizing using PIL.Image. We use default pyiqa=0.1.15.post2 to calculate metrics.
To ablate the effect of resizing, we evaluate the same 28-step predictions on all 115 LOL images under two protocols: (A) native 1024 × 1024 outputs against bicubic-upsampled GT, (B) outputs bicubic-resized to the original GT dimensions against unchanged GT.
Compared to resized evaluation, our original resolution evaluation leads to PSNR and SSIM generally decreases but LPIPS [7] all improves. These differences reflect resampling and evaluation-scale effects. We have clarified this protocol in the appendix.
| Metric | A: 1024 × 1024 | B: Original GT dimensions | Δ (B-A) | Images improved |
|---|---|---|---|---|
| RGB PSNR ↑ (dB) | 21.34 | 21.29 | -0.050 | 45/115 |
| SSIM-YIQ ↑ | 0.858 | 0.838 | -0.020 | 86/115 |
| LPIPS-AlexNet ↓ | 0.230 | 0.159 | −0.071 | 115/115 |
Q5: Generalization to dehazing and underwater enhancement.
Thank you for this constructive suggestion. We have added quantitative and qualitative results in Appendix A.9, and copy the main results here. These tasks share the challenge of recovering scene content under degraded visibility, while scattering and wavelength-dependent attenuation require task-specific adaptation.
Evaluation setting. We evaluate the our UG-Prior model on fixed exploratory subsets of 16 O-HAZE images and 30 dark UIEB images, without task-specific fine-tuning.
Dehazing. We equipped UG-Prior with WB and DCP as preprocessor to adapt the frozen low-light enhancement model to accept hazy inputs. We evaluate our method and three task-specific methods (DehazeFormer, DEA-Net, DNMGDT) with three full-reference metrics and four non-reference metrics. The comparisons use the official outdoor DehazeFormer-B and DEA-Net checkpoints and the released DNMGDT checkpoint. The qualitative and quantitative comparison are shown below:
| Method | NIQE ↓ | BRISQUE ↓ | MUSIQ ↑ | CLIP-IQA ↑ | PSNR ↑ (dB) | SSIM ↑ | LPIPS ↓ |
|---|---|---|---|---|---|---|---|
| DehazeFormer, TIP'23 | 2.187 | 8.307 | 66.515 | 0.3682 | 17.326 | 0.7058 | 0.3057 |
| DEA-Net, TIP'24 | 2.197 | 9.162 | 66.831 | 0.3935 | 17.199 | 0.7185 | 0.2958 |
| DNMGDT, TMM'25 | 4.860 | 34.522 | 53.816 | 0.3508 | 18.333 | 0.7070 | 0.4291 |
| Ours | 2.194 | 10.345 | 64.433 | 0.4512 | 18.003 | 0.7539 | 0.2875 |
We provide additional visualized results and limitation analysis in Appendix A.9 Fig.20 and Fig.21.
Underwater enhancement. We evaluate the frozen UG-Prior model without underwater-specific fine-tuning on a fixed 30-image UIEB subset. We compare against three underwater-trained models including QualiTeacher (AIM-Net-based) [ECCV'26] [11] PUIE-Net [ECCV'22] [12] and NU2Net [AAAI 2023] [13] using its official pretrained underwater checkpoint and inference procedure.
| Method | LPIPS ↓ | NIQE ↓ | BRISQUE ↓ | MUSIQ ↑ | CLIP-IQA ↑ |
|---|---|---|---|---|---|
| PUIE-Net, ECCV'22 | 0.1702 | 4.3507 | 30.5922 | 43.0376 | 0.2827 |
| NU2Net, AAAI'23 | 0.2679 | 4.7923 | 38.9131 | 36.7245 | 0.2733 |
| QualiTeacher, ECCV'26 | 0.2551 | 4.2410 | 25.1324 | 56.0311 | 0.3960 |
| UG-Prior | 0.3011 | 3.4606 | 13.3464 | 55.5135 | 0.4346 |
UG-Prior achieves the best on three of the four no-reference metrics and the second best on the remaining one.
Specifically, it obtains better per-image scores on 26/30 images for NIQE, 26/30 for BRISQUE, 15/30 for MUSIQ, and 21/30 for CLIP-IQA.
Although visually plausible, as UG-Prior use no paired UIE data during training, it is hard to establish superior restoration fidelity required by the exact ground-turth images. We provide visualized comparison in Appendix A.9 Fig.22 and discuss more, verifying the universality of our method.
Official Response to Reviewer mrSq (Part 3/3)
References
- Wenjing Wang et al. Zero-Reference Low-Light Enhancement via Physical Quadruple Priors. CVPR, 2024.
- Chen Chen et al. Learning to See in the Dark. CVPR, 2018. SID dataset.
- Chen Wei et al. Deep Retinex Decomposition for Low-Light Enhancement. BMVC, 2018. LOL dataset.
- Ben Fei et al. Generative Diffusion Prior for Unified Image Restoration and Enhancement. CVPR, 2023.
- Huaqiu Li et al. LD-RPS: Zero-Shot Unified Image Restoration via Latent Diffusion Recurrent Posterior Sampling. ICCV, 2025.
- Zhou Wang et al. Image Quality Assessment: From Error Visibility to Structural Similarity. IEEE Transactions on Image Processing, 2004.
- Richard Zhang et al. The Unreasonable Effectiveness of Deep Features as a Perceptual Metric. CVPR, 2018.
- Codruta O. Ancuti et al. O-HAZE: A Dehazing Benchmark With Real Hazy and Haze-Free Outdoor Images. CVPR Workshops, 2018.
- Chongyi Li et al. An Underwater Image Enhancement Benchmark Dataset and Beyond. IEEE Transactions on Image Processing, 2019.
- Kaiming He, Jian Sun, and Xiaoou Tang. Single Image Haze Removal Using Dark Channel Prior. CVPR, 2009.
- Fengyang Xiao et al. QualiTeacher: Quality-Conditioned Pseudo-Labeling for Real-World Image Restoration. ECCV, 2026.
- Zhenqi Fu et al. Uncertainty Inspired Underwater Image Enhancement. ECCV, 2022.
- Chunle Guo et al. Underwater Ranker: Learn Which Is Better and How to Be Better. AAAI, 2023.
- Yuda Song et al. Vision Transformers for Single Image Dehazing. IEEE TIP, 2023.
- Zixuan Chen et al. DEA-Net: Single Image Dehazing Based on Detail-Enhanced Convolution and Content-Guided Attention. IEEE TIP, 2024.
- Yanzhao Su et al. Real Scene Single Image Dehazing Network With Multi-Prior Guidance and Domain Transfer. IEEE TMM, 2025.
Official Response to Reviewer mrSq (Part 1/3)
We appreciate the reviewer’s positive assessment and constructive suggestions for directly validating uncertainty–error correspondence, the accuracy of online prior refinement, and broader generalization capabilties on dehazing/under water enhancement.
Q1 / Weakness 1: Direct evidence to verify the correspondence between uncertainty and prior error.
We provide additional analysis demonstrating the correspondence between estimated uncertainty and empirical prior error. As described in our response to
R1-Q1, we evaluate 291 short–long exposure pairs from 49 SID Fuji scenes [2], with three short exposures (33/40/100 ms), five Gaussian scales, and three patch sizes (8/16/32). At each scale, prior error is computed as the MSE between the patch-normalized short- and long-exposure priors. The matched long exposure provides the empirical reference prior. Uncertainty is computed from the short exposure images alone.Because uncertainty is u=1−q, its correlation with error has the opposite sign to the confidence–error correlation. All 45 scene-averaged exposure–scale–patch-size correlations are positive. Representative Pearson coefficients are shown below. Each cell reports the fine/coarse scales, σ=0.5/2.5.
These results directly support uncertainty as an empirical reliability cue, with a stronger association at fine scales. Therefore, downweighting fine-scale responses in unreliable regions while retaining their detail in reliable regions intuitively improves the robustness–informativeness trade-off, as shown in Figure 3. We provide a more straightforward figure in Appendix A.1.
Q2 / Weaknesses 2: More direct evidence for prior refinement mechanism.
To explore whether prior representations gradually become more accurate with iteration, we record the evolving prior at every flow timestep on 15 LOL-v1 test images over 28 sampling steps [3]. Its error is measured against the normal-light reference prior and divided by the initial prior error for each image.
The table reports mean relative prior MSE ↓. Note that the relative error remains 1 during steps 1–8 because the first updated prior is used at step 9. The final optimized prior has 33.8% lower error than the initial prior, with improvement on all 15 images. This experiment provides further evidence for the effectiveness of the proposed online prior–image co-optimization mechanism.
Compared to general iterative prior correction methods, our methodological distinction lies in the learned interaction between image generation and prior refinement, rather than the fusion formula itself. Our mechanism refines the physical conditioning prior through bidirectional interaction with the evolving generative features with all network weights fixed and no test-time backpropagation. This differs from test-time optimization in existing generative restoration methods. We have included this experiment in
Sec.4.5, Page 10.Q3: Were the backbone and training settings controlled in Table 3?
Yes. The 21.29-versus-19.10 dB comparison uses the same FLUX generation backbone, training data, and training hyperparameters, exclusively replacing the conditioning prior with QuadPrior's [1]. We have clarified this in the main text. This result reveals a counterintuitive phenomenon: a stronger backbone does not necessarily compensate for a less suitable condition. One possible explanation is that a high-capacity generator learns perturbation-specific mappings when the prior inadequately balances robustness and information retention. In contrast, our prior enables more effective conditioning under the controlled setting.