Research map / Vision-language and multimodal / Human and embodied modeling
A Dataset of Flash and Ambient Illumination Pairs from the Crowd
Yagiz Aksoy, Changil Kim, Petr Kellnhofer, Sylvain Paris, Mohamed Elgharib, Marc Pollefeys, Wojciech Matusik
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 |
|---|---|---|---|
| add L2 loss E to ratio loss R (ablation) | Pix2Pix ratio-image network · Ambient illumination estimation (alpha=0.5) | PSNR: 19.199 -> 20.261 | Improved |
| add ratio loss R to L2 loss E (ablation) | Pix2Pix ratio-image network · Ambient illumination estimation (alpha=0.5) | PSNR: 19.333 -> 20.261 | Improved |
| add L2 loss E to ratio loss R (ablation) | Pix2Pix ratio-image network · Ambient illumination estimation (alpha=0.5) | SSIM: 0.707 -> 0.804 | Improved |
| add ratio loss R to L2 loss E (ablation) | Pix2Pix ratio-image network · Ambient illumination estimation (alpha=0.5) | SSIM: 0.716 -> 0.804 | Improved |
| add perceptual loss P to RE (ablation) | Pix2Pix ratio-image network · Ambient illumination estimation (alpha=0.5) | PSNR: 20.261 -> 19.568 | Hurt |
| add perceptual loss P to RE (ablation) | Pix2Pix ratio-image network · Ambient illumination estimation (alpha=0.5) | SSIM: 0.804 -> 0.773 | Hurt |
| add adversarial loss A to RE (ablation) | Pix2Pix ratio-image network · Ambient illumination estimation (alpha=0.5) | PSNR: 20.261 -> 18.116 | Hurt |
| add adversarial loss A to RE (ablation) | Pix2Pix ratio-image network · Ambient illumination estimation (alpha=0.5) | SSIM: 0.804 -> 0.495 | Hurt |
| add adversarial loss A to RE (ablation) | Pix2Pix ratio-image network · Ambient illumination estimation (alpha=0.5) | PSNR (alpha=0.1): 15.220 -> 14.099 | Hurt |
| add L2 loss E to ratio loss R (ablation) | Pix2Pix ratio-image network · Ambient illumination estimation (alpha=0.5) | PSNR (alpha=0.1): 14.801 -> 15.220 | Improved |
| predict ratio image and apply domain-transform filtering instead of direct illumination estimation | Pix2Pix U-net · Ambient illumination estimation (alpha=0.5) | qualitative artifacts | Improved |
Closest papers by meaning
- Illuminant Spectra-Based Source Separation Using Flash Photography (CVPR 2018 · similarity 0.86)
- Computational Flash Photography Through Intrinsics (CVPR 2023 · similarity 0.86)
- A Dataset of Multi-Illumination Images in the Wild (ICCV 2019 · similarity 0.86)
- SunFaded: Illumination-Aware Gaussian Splatting for Dark Scenes with Camera-Mounted Active Lighting (CVPR 2026 · similarity 0.85)
- Flash-Split: 2D Reflection Removal with Flash Cues and Latent Diffusion Separation (CVPR 2025 · similarity 0.84)
- Towards Robust Event-guided Low-Light Image Enhancement: A Large-Scale Real-World Event-Image Dataset and Novel Approach (CVPR 2024 · 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.
