Research map / Video / Video generation
Video Color Grading via Look-Up Table Generation
Seunghyun Shin, Dongmin Shin, Jisu Shin, Hae-Gon Jeon, Joon-Young Lee
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 |
|---|---|---|---|
| generate LUT explicitly with diffusion instead of implicit LUT with perceptual loss (ablation) | Video color grading (L-Diffuser) · Condensed Movie (LUT-graded test) | PSNR: 20.87 -> 24.55 | Improved |
| generate LUT explicitly with diffusion instead of implicit LUT with l2 loss (ablation) | Video color grading (L-Diffuser) · Condensed Movie (LUT-graded test) | PSNR: 22.17 -> 24.55 | Improved |
| use L-Diffuser instead of LUT bases-based generation (ablation) | Video color grading (L-Diffuser) · Condensed Movie (LUT-graded test) | LPIPS: 0.3562 -> 0.1457 | Improved |
| use L-Diffuser instead of MLP-based LUT generation (ablation) | Video color grading (L-Diffuser) · Condensed Movie (LUT-graded test) | PSNR: 19.73 -> 24.55 | Improved |
| use L-Diffuser instead of NLUT with explicit LUT (ablation) | Video color grading (L-Diffuser) · Condensed Movie (LUT-graded test) | PSNR: 13.97 -> 24.55 | Improved |
| use GS-Extractor instead of color histogram features (ablation) | Video color grading (L-Diffuser) · Condensed Movie (LUT-graded test) | PSNR: 18.52 -> 24.55 | Improved |
| use GS-Extractor instead of VGG features (ablation) | Video color grading (L-Diffuser) · Condensed Movie (LUT-graded test) | PSNR: 18.95 -> 24.55 | Improved |
| generate LUT instead of directly generating frames with image editor (ablation) | Video color grading (L-Diffuser) · Condensed Movie (LUT-graded test) | BRISQUE: 53.39 -> 40.23 | Improved |
| select key-frame pairs with CLIP instead of random (ablation) | Video color grading (L-Diffuser) · Condensed Movie (LUT-graded test) | PSNR: 23.63 -> 24.55 | Improved |
| use CLIP instead of ResNeXt for key-frame selection (ablation) | Video color grading (L-Diffuser) · Condensed Movie (LUT-graded test) | PSNR: 24.54 -> 24.55 | No clear effect |
| apply user text-prompt feedback retouching | Video color grading · User Study 2 | preference score (1-5): 3.64 | Improved |
Closest papers by meaning
- CanonCGT: Reference-Based Color Grading via Canonical Pivot Representation (CVPR 2026 · similarity 0.87)
- Automatic Controllable Colorization via Imagination (CVPR 2024 · similarity 0.84)
- Image Referenced Sketch Colorization Based on Animation Creation Workflow (CVPR 2025 · similarity 0.83)
- Efficient Neural Network Encoding for 3D Color Lookup Tables (AAAI 2025 · similarity 0.83)
- ChromaFusionNet (CFNet): Natural Fusion of Fine-Grained Color Editing (AAAI 2024 · similarity 0.83)
- MangaNinja: Line Art Colorization with Precise Reference Following (CVPR 2025 · similarity 0.82)
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.
