Research map / Image generation / Image restoration
Robust Reflection Removal With Reflection-Free Flash-Only Cues
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 reflection-free flash-only cue with the dedicated reflection-then-transmission architecture | Reflection removal · Real-world dataset | PSNR: 24.53 -> 29.76 | Helped |
| Use the reflection-free flash-only cue with the dedicated reflection-then-transmission architecture | Reflection removal · Real-world dataset | SSIM: 0.890 -> 0.930 | Helped |
| Use the reflection-free flash-only cue with the dedicated reflection-then-transmission architecture | Reflection removal · Real-world dataset | LPIPS: 0.205 -> 0.156 | Helped |
| Add flash-only image Ifo rather than using only ambient image Ia (ablation) | gR + gT reflection removal · Real-world dataset | PSNR: 25.13 -> 29.76 | Helped |
| Add flash-only image Ifo rather than using only ambient image Ia (ablation) | gR + gT reflection removal · Real-world dataset | SSIM: 0.888 -> 0.930 | Helped |
| Add flash-only image Ifo rather than using only ambient image Ia (ablation) | gR + gT reflection removal · Real-world dataset | LPIPS: 0.258 -> 0.156 | Helped |
| Use flash-only image Ifo instead of flash image If (ablation) | gB direct transmission model · Real-world dataset | PSNR: 26.99 -> 27.55 | Helped |
| Use flash-only image Ifo instead of flash image If (ablation) | gR + gT reflection removal · Real-world dataset | PSNR: 27.21 -> 29.76 | Helped |
| Estimate reflection first and then estimate transmission using the ambient image and estimated reflection, rather than directly estimating transmission (ablation) | Reflection removal with Ia and Ifo · Real-world dataset | PSNR: 27.55 -> 29.76 | Helped |
| Estimate reflection first and then estimate transmission using the ambient image and estimated reflection, rather than directly estimating transmission (ablation) | Reflection removal with Ia and Ifo · Real-world dataset | LPIPS: 0.187 -> 0.156 | Helped |
Closest papers by meaning
- Flash-Split: 2D Reflection Removal with Flash Cues and Latent Diffusion Separation (CVPR 2025 · similarity 0.91)
- PolarFree: Polarization-based Reflection-Free Imaging (CVPR 2025 · similarity 0.89)
- Revisiting Single Image Reflection Removal In the Wild (CVPR 2024 · similarity 0.88)
- FIRM: Flexible Interactive Reflection ReMoval (AAAI 2025 · similarity 0.87)
- Removing Reflections from RAW Photos (CVPR 2025 · similarity 0.86)
- Polarization Guided Mask-Free Shadow Removal (AAAI 2025 · similarity 0.83)
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.