- Rating
- 2(+1)
- Confidence
- 4
- Informativeness
- 1
- Presentation
- 2
- Soundness
- 1
Read review
Paper summary
This paper addresses real‑world image demoiréing by leveraging a pre‑trained large generative model. Novel strategies are proposed for both training and inference, yielding improved subjective scores over several general‑purpose and task‑specific demoiréing methods, on both the self‑constructed no‑reference test set and two public benchmarks. Nevertheless, the method achieves slightly lower full‑reference fidelity (PSNR / SSIM) compared with pixel‑wise regression‑based approaches.
Critical strengths
- The newly introduced dataset consists of real‑world acquisitions from various mainstream smartphones. Further scaling and better alignment with real‑user data distribution would make it a valuable asset for the community.
- The integration of condition‑shift in Flow‑Matching image editing and dual‑scale inference is well‑tailored to moiré‑specific degradation. Driven by clear motivations, the method offers a targeted solution to real-world demoiréing challenges.
Critical weaknesses
Despite the authors’ emphasis on subjective evaluation as the key evidence for the superiority of their approach over competing methods, I still have concerns regarding the quality of the reported results:
- The proposed method yields generally inferior pixel‑wise metrics compared with competing baselines. While the authors argue that higher PSNR of other methods stems from "a trade‑off between pixel fidelity and artifact removal", it does not fully rule out potential detail degradation and undesired content distortion in model outputs.
- The baseline results reported on FHDMi and UHDM deviate from the numbers originally published in their respective papers. It appears that retraining these baselines under the same setup may degrade baseline performance, which could lead to unfair comparisons. Moreover, the paper lacks comparisons against Freqformer — another dual‑scale demoiréing approach — because its code and weights were unavailable at submission time. According to the original Freqformer paper, it achieves competitive metrics with substantially lower computational overhead.
- Visually, noticeable color shifts can be observed in most examples in Figure 6, and observable loss of fine‑grained details appears in Figure 5.
Questions for authors
- I'm wondering the motivation of establishing a newly‑built DailyMoiré130 test set: although the paper argues existing benchmarks suffer from limited conditions and scenes, I remain unconvinced that merely 130 images are sufficient to represent the data distribution encountered under realistic scenarios. I would appreciate further explanations on the sampling strategy and statistical properties of this dataset, to justify why such a small‑scale collection can reliably evaluate real‑world generalization.
- Why are the objective metrics (PSNR, etc.) for retrained baselines degrading? This is the most confusing problem for me. Ideally, comparisons should be performed using officially released baseline weights. If the authors choose to retrain all baselines from scratch, thorough analysis must be provided to pinpoint where this performance drop originates and rule out unfair comparison.
- The paper only ablates hard binary condition‑shifting with different r values (), but does not compare against the more classical strategy of continuously interpolating between the two conditioning latents over denoising timesteps, which is a well‑established scheduling paradigm within diffusion and Flow‑Matching image editing. It raises the question: can comparable performance be achieved with continuous weighted scheduling? If interpolation yields similar results, the necessity of the hard‑shift design would be weakened. I suggest the authors to add this ablation and corresponding analysis.
7 comments
Response to Reviewer NgSt (W2, Q2)
W2 and Q2 Retraining and official checkpoints
We have supplemented the FHDMi and UHDM comparisons with official ESDNet-L checkpoints to address the concern about relying on retrained baselines (Table 4). We also conducted additional evaluations on DailyMoiré130 to assess cross-domain generalization under unconstrained real-world capture conditions (Tables 5 and 6).
The PSNR drop is consistent with a trade-off between single-domain specialization and multi-domain generalization. Official ESDNet-L checkpoints are trained on individual benchmarks, whereas our unified training shares model capacity across 12 heterogeneous streams, including eight public or synthetic moiré streams, two document streams, and two self-reconstruction streams (Appendix C.1).
Learning across these domains may reduce specialization to benchmark-specific pixel statistics. On unseen DailyMoiré130 captures, retrained ESDNet-L nevertheless achieves lower RMC and RMA than the official UHDM-trained checkpoint (Table 5), while CLIP-IQA+ is slightly lower.
In our human evaluation across all 130 images (Table 6), retrained ESDNet-L is preferred nearly twice as often as the official checkpoint across moiré removal (96 vs. 51), content preservation (91 vs. 46), and overall quality (94 vs. 51), achieving GSB scores of 1.41 to 1.43. This confirms that mixed training strengthens rather than degrades baseline performance on unconstrained real-world captures.
Crucially, our single unified model reduces RMC by 55.2% on FHDMi and 63.5% on UHDM against the official ESDNet-L checkpoints, with higher MUSIQ and CLIP-IQA+ on both datasets. TOPIQ is higher on FHDMi and tied on UHDM (Table 4). On DailyMoiré130, RMC and RMA are reduced by 67.3% and 66.6%, with higher CLIP-IQA+ (Table 5). Separately, the Best-of-7 study gives our method 71.3% overall quality preference versus 8.8% for retrained ESDNet-L.
These results support the generalization of our framework to challenging in-the-wild moiré. Aligned visual comparisons against official ESDNet-L are included in the supplementary material (Appendix L, manuscript Figure 12).
Table 4. Official and retrained comparison on paired benchmarks
| Dataset | Method | PSNR ↑ | MUSIQ ↑ | CLIP-IQA+ ↑ | TOPIQ ↑ | RMC ↓ |
|---|---|---|---|---|---|---|
| FHDMi | Official ESDNet-L | 24.88 | 58.31 | 0.542 | 0.477 | 0.0143 |
| FHDMi | Retrained ESDNet-L | 20.67 | 55.56 | 0.523 | 0.408 | 0.0175 |
| FHDMi | Ours | 21.09 | 61.55 | 0.585 | 0.508 | 0.0064 |
| UHDM | Official ESDNet-L | 22.42 | 65.58 | 0.643 | 0.585 | 0.0107 |
| UHDM | Retrained ESDNet-L | 21.16 | 65.44 | 0.625 | 0.556 | 0.0139 |
| UHDM | Ours | 19.99 | 67.43 | 0.656 | 0.585 | 0.0039 |
TOPIQ denotes the no-reference metric, and official ESDNet-L checkpoints are evaluated under the same test protocol as our main comparisons.
Table 5. DailyMoiré130 objective comparison
| Method | RMC ↓ | RMA ↓ | CLIP-IQA+ ↑ |
|---|---|---|---|
| Official ESDNet-L | 0.0352 | 0.1652 | 0.5517 |
| Retrained ESDNet-L | 0.0340 | 0.1599 | 0.5478 |
| Ours | 0.0115 | 0.0551 | 0.5928 |
Table 6. Retrained against official ESDNet-L
| Criterion | Good retrained | Same | Bad retrained | GSB ↑ |
|---|---|---|---|---|
| Moiré removal | 96 (37.65%) | 108 (42.35%) | 51 (20.00%) | 1.43 |
| Content preservation | 91 (35.69%) | 118 (46.27%) | 46 (18.04%) | 1.43 |
| Overall quality | 94 (36.86%) | 110 (43.14%) | 51 (20.00%) | 1.41 |
This evaluation contains 255 image pair judgments from 15 participant IDs covering all 130 images. GSB = (G + 0.5S)/(B + 0.5S). Good and Bad refer to the retrained checkpoint. Values above 1 favor retraining.
Released ESDNet-L checkpoints are marked with a superscript in manuscript Table 2. For completeness, Table 7 lists the published results of MZNet-Large [1] and Freqformer [2] (whose code and weights remain unavailable) as reported in their original papers.
Table 7. Original paper results for literature reference
| Method | Dataset | PSNR ↑ | SSIM ↑ | LPIPS ↓ | Source |
|---|---|---|---|---|---|
| MZNet-Large | FHDMi | 26.12 | 0.8624 | 0.1042 | [1] Table 6 |
| MZNet-Large | UHDM | 23.63 | 0.8096 | 0.2237 | [1] Table 6 |
| Freqformer | FHDMi | 25.26 | 0.8518 | 0.1253 | [2] Tables 1 and 2 |
| Freqformer | UHDM | 22.24 | 0.8021 | 0.2424 | [2] Tables 1 and 2 |
Results are cited directly from the original papers [1] and [2] under their respective benchmark-specific training and evaluation protocols.
References
[1] Lee et al. Moiré Zero. An Efficient and High Performance Neural Architecture for Moiré Removal. arXiv 2507.22407.
[2] Liu et al. Freqformer. Image Demoiréing Transformer via Effective Frequency Decomposition. arXiv 2505.19120.
Response to Reviewer NgSt (W3, Q1)
W3 Color shifts and fine details
We evaluate color accuracy and detail fidelity using GT-referenced metrics and human judgments to examine how local color shifts and fine-detail loss affect restoration quality (Tables 1 to 3). The results show that our method achieves GT color errors comparable to ESDNet-L and competitive detail preservation overall.
Regarding color shifts, we agree that some moiré-affected regions can appear lighter or whiter after restoration. This behavior can be understood in terms of the pretrained natural image prior. When severe corruption obscures the underlying colors, the pretrained prior may favor a clean, uniform appearance while removing the corruption.
Importantly, a color change relative to the corrupted input does not necessarily imply a larger error relative to the clean GT. Our UHDM chroma and hue errors are comparable to ESDNet-L (Table 3, Appendix G, manuscript Table 13).
Table 3. Color fidelity on UHDM
| Method | Mean absolute chroma error ↓ | Mean hue error ↓ |
|---|---|---|
| ESDNet-L | 4.454 | 2.776 |
| Ours | 4.422 | 2.643 |
Regarding fine details, we also acknowledge that fine details can be lost in individual cases. However, the apparent detail loss in Figure 5 should be considered together with the removal of severe moiré. We therefore provide dedicated detail preservation judgments and GT referenced perceptual evaluation to evaluate content fidelity alongside artifact removal.
Our method receives 68.5% of the detail preservation selections and performs competitively on TOPIQ-FR, PieAPP, and AHIQ (Tables 1 and 2). We discuss the restoration and fidelity tradeoff in Sec. 5 and have made the possibility of fine detail loss explicit.
Q1 Dataset motivation, sampling, and statistical properties
Our motivation is to evaluate challenging moiré on unconstrained in-the-wild displays, particularly those in malls and stations, where fine stripes and broad color bands often coexist. These conditions are underrepresented in the designated monitor settings of TIP2018, FHDMi, and UHDM. Yue et al. [1] similarly note that large public screens exhibit more visible moiré while lacking accessible clean ground truth.
We photographed displays operating in unconstrained real-world environments using five smartphone models at ordinary viewing positions and selected captures with visibly apparent moiré. The collection covers four scene categories (Tables 8 and 9). This curated sampling specifically targets hard, mixed-frequency real-world cases, complementing large-scale controlled monitor benchmarks (FHDMi and UHDM) while remaining compact enough for dense human evaluation across all 130 images (558 total judgments).
Table 8. Benchmark capture conditions
| Dataset | Screen types | Capture settings | Capture devices | Image resolution |
|---|---|---|---|---|
| TIP2018 | Designated monitors and laptop screens | Controlled screen recapture | iPhone 6, Samsung Galaxy S7 Edge, Sony Xperia Z5 Premium Dual | 415p to 968p |
| FHDMi | Designated monitors and laptop screens | Controlled screen recapture | iPhone 7, Samsung Galaxy S10, Huawei Mate 20 | 1080p |
| UHDM | Designated desktop monitors | Controlled screen recapture | iPhone XR, iPhone 13, Redmi K30 Pro | 2160p |
| DailyMoiré130 | Commercial signage, public display walls, monitors, and laptops | Unconstrained in-the-wild settings | vivo X200s, iPhone 17, iPhone 16 Pro Max, HONOR 90 Pro, Huawei Enjoy 90 Pro Max | 864p to 4284p |
Table 9. Overall quality preferences by scene
| Scene | Images | Judgments | Ours | ESDNet-L | Uni-control | MZNet | OSEDiff | MoCE-IR | AdaIR | No clear best |
|---|---|---|---|---|---|---|---|---|---|---|
| Public or commercial indoor | 86 | 355 | 73.8% | 7.6% | 6.2% | 2.5% | 2.0% | 2.5% | 2.3% | 3.1% |
| Outdoor public | 18 | 85 | 58.8% | 10.6% | 5.9% | 2.4% | 7.1% | 3.5% | 3.5% | 8.2% |
| Transportation | 13 | 59 | 79.7% | 3.4% | 6.8% | 1.7% | 3.4% | 0.0% | 0.0% | 5.1% |
| Office or home | 13 | 59 | 66.1% | 18.6% | 10.2% | 3.4% | 0.0% | 0.0% | 0.0% | 1.7% |
Selection rates are percentages of all judgments within each scene, including no clear best. Scene proportions are 66.15%, 13.85%, 10.00%, and 10.00%.
We additionally stratified the Best-of-7 study by scene. The 31 participants provide 558 overall quality judgments covering all 130 images. Our method receives the highest preference rate in every scene category (Table 9). Moiré suppression and detail preservation show the same leading method across all four categories.
These results support generalization across the included in-the-wild capture conditions and complement paired evaluation on FHDMi and UHDM. We have included the complete scene statistics and all three dimensions of scene stratified human evaluation in the supplementary material.
References
[1] Yue et al. Unsupervised moiré pattern removal for recaptured screen images. Neurocomputing, 2021.
Response to Reviewer NgSt (W1)
We thank Reviewer NgSt for the valuable feedback and detailed suggestions. Below we address each of the concerns and questions in detail.
W1 Pixel fidelity and content preservation
To examine whether lower PSNR reflects content degradation, we evaluate detail and structure preservation using dedicated human judgments and GT-referenced perceptual metrics, which support competitive content fidelity overall (Tables 1 and 2).
In the DailyMoiré130 Best-of-7 study, participants assess text, edges, and detail preservation separately from moiré suppression. Our method receives 68.5% of the 558 detail preservation selections from 31 participants (Table 1, Appendix E). On UHDM, TOPIQ-FR and AHIQ measure perceptual quality, while PieAPP measures perceptual error.
Our method achieves the highest TOPIQ-FR and the second best PieAPP and AHIQ among the compared methods (Table 2, Appendix G, manuscript Table 13). These complementary evaluations support content preservation beyond pixelwise agreement.
Table 1. Detail preservation on DailyMoiré130
Table 2. GT referenced perceptual evaluation on UHDM