Research map / Image generation / Image restoration
Self-Adaptively Learning to Demoiré from Focused and Defocused Image Pairs
Lin Liu, Shanxin Yuan, Jianzhuang Liu, Liping Bao, Gregory Slabaugh, Qi Tian
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use the focused moiré image rather than a concatenation of focused and defocused images as D's input (ablation) | FDNet · SynScreenMoire | PSNR: 27.43 -> 31.77 | Improved |
| Use the focused moiré image rather than random noise as D's input (ablation) | FDNet · SynScreenMoire | PSNR: 28.42 -> 31.77 | Improved |
| Use the focused moiré image rather than only the defocused image as D's input (ablation) | FDNet · SynScreenMoire | PSNR: 27.17 -> 31.77 | Improved |
| Use end-to-end image prediction instead of predicting linear parameters for the defocused image (ablation) | FDNet · SynScreenMoire | PSNR: 29.57 -> 31.77 | Improved |
| Jointly optimize the demoiréing and blur-kernel networks instead of alternating their updates (ablation) | FDNet · SynScreenMoire | PSNR: 27.31 -> 31.77 | Improved |
| Use a blur-kernel generator network instead of a directly learnable kernel (ablation) | FDNet · SynScreenMoire | PSNR: 26.10 -> 31.77 | Improved |
| Use a U-Net rather than an encoder-decoder architecture (ablation) | FDNet · SynScreenMoire | PSNR: 0.11dB | Improved |
| Use a U-Net rather than a ResNet architecture (ablation) | FDNet · SynScreenMoire | PSNR: 1.05dB | Improved |
| Use FDNet rather than the strongest reported baseline | Demoiréing methods · SynScreenMoire | PSNR: 31.06 -> 31.77 | Improved |
| Use FDNet rather than the strongest reported baseline | Demoiréing methods · SynTextureMoire | PSNR: 24.70 -> 25.98 | Improved |
| Use FDNet rather than the strongest baseline | Demoiréing methods · RealScreenMoire | NIQE: 5.34 -> 5.11 | Improved |
| Use FDNet rather than the strongest baseline | Demoiréing methods · RealTextureMoire | BRISQUE: 42.13 -> 41.77 | Improved |
Closest papers by meaning
- Coarse-to-fine Disentangling Demoiréing Framework for Recaptured Screen Images (TPAMI 2023 · similarity 0.91)
- Video Demoireing Using Focused-Defocused Dual-Camera System (TPAMI 2025 · similarity 0.90)
- Learning Image Demoiréing from Unpaired Real Data (AAAI 2024 · similarity 0.90)
- UniDemoiré: Towards Universal Image Demoiréing with Data Generation and Synthesis (AAAI 2025 · similarity 0.87)
- Direction-Aware Video Demoiréing with Temporal-Guided Bilateral Learning (AAAI 2024 · similarity 0.87)
- Through the Lens: Benchmarking Deepfake Detectors Against Moiré-Induced Distortions (NeurIPS 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
