Research map / Image generation / Text-to-image diffusion
RawGen: Learning Camera Raw Image Generation
Dongyoung Kim, Junyong Lee, Abhijith Punnappurath, Mahmoud Afifi, Sangmin Han, Alex Levinshtein, Michael S Brown
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 |
|---|---|---|---|
| train with many-to-one supervision (multiple sRGB renditions to one XYZ anchor) instead of one-to-one (ablation) | RawGen · MIT-Adobe FiveK Expert A | PSNR: 19.57 -> 23.20 | Improved |
| train with many-to-one supervision instead of one-to-one (ablation) | RawGen · MIT-Adobe FiveK Expert E | PSNR: 18.07 -> 23.89 | Improved |
| use RawGen instead of CIE XYZ Net for sRGB-to-XYZ under unseen retouching styles | inverse ISP · MIT-Adobe FiveK Expert B | PSNR: 21.04 -> 24.35 | Improved |
| train illuminant estimator on RawGen text-to-raw data instead of Graphics2RAW synthetic data | illuminant estimator · NUS-8 | mean angular error: 4.21 -> 3.14 | Improved |
| train neural ISP on RawGen synthetic raw instead of Graphics2RAW data | neural ISP · Neural ISP (night dataset) | PSNR: 38.10 -> 38.42 | Improved |
| train neural ISP on RawGen synthetic raw instead of Graphics2RAW data | neural ISP · Neural ISP (night dataset) | SSIM: 0.974 -> 0.970 | Hurt |
| train denoiser on RawGen synthetic clean images instead of Graphics2RAW data | raw denoiser · Raw denoising ISO 1600 | PSNR: 49.37 -> 50.63 | Improved |
| train denoiser on RawGen synthetic data instead of Graphics2RAW | raw denoiser · Raw denoising ISO 3200 | PSNR: 48.16 -> 48.57 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- CIE XYZ Net — CIE XYZ Net: Unprocessing Images for Low-Level Computer Vision Tasks (TPAMI 2021) · better on 4 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Generative Photography: Scene-Consistent Camera Control for Realistic Text-to-Image Synthesis (CVPR 2025 · similarity 0.87)
- ReRAW: RGB-to-RAW Image Reconstruction via Stratified Sampling for Efficient Object Detection on the Edge (CVPR 2025 · similarity 0.86)
- More Than Generation: Unifying Generation and Depth Estimation via Text-to-Image Diffusion Models (NeurIPS 2025 · similarity 0.85)
- GenFusion: Closing the Loop between Reconstruction and Generation via Videos (CVPR 2025 · similarity 0.85)
- DiffuseHigh: Training-Free Progressive High-Resolution Image Synthesis Through Structure Guidance (AAAI 2025 · similarity 0.85)
- DiffRAW: Leveraging Diffusion Model to Generate DSLR-Comparable Perceptual Quality sRGB from Smartphone RAW Images (AAAI 2024 · similarity 0.85)
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.
