Research map / Image generation / Image restoration
Real-Time Image Enhancer via Learnable Spatial-Aware 3D Lookup Tables
Tao Wang, Yong Li, Jingyang Peng, Yipeng Ma, Xian Wang, Fenglong Song, Youliang Yan
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 |
|---|---|---|---|
| increase basic LUTs from 3 to 30 in original 3DLUT (ablation) | 3DLUT · HDR+ 480p (ours) | PSNR: 19.91 -> 20.29 | Improved |
| replace the original predictor with our predictor (1D weight only, 30 LUTs) (ablation) | 3DLUT(30,0) · HDR+ 480p (ours) | PSNR: 20.29 -> 20.38 | Improved |
| add 3D pixel-wise weight map with spatial-aware LUTs (ablation) | Ours(30,0) · HDR+ 480p (ours) | PSNR: 20.38 -> 22.52 | Improved |
| add 1D scenario weight (T=3) to the 3D weight head, M=10 (ablation) | Ours(0,30) · HDR+ 480p (ours) | PSNR: 22.52 -> 22.73 | Improved |
| full two-head spatial-aware model versus original 3DLUT (ablation) | Ours(3,10) · HDR+ 480p (ours) | PSNR: 19.91 -> 22.73 | Improved |
| increase number of categories M from 1 up to 10 (ablation) | spatial-aware 3D LUTs · HDR+ 480p (ours) | PSNR | Improved |
| increase M beyond 10 (ablation) | spatial-aware 3D LUTs · HDR+ 480p (ours) | PSNR | No clear effect |
| run at 4K full resolution instead of 480p (ablation) | Ours(3,10) · HDR+ ablation | PSNR: 22.73 -> 22.56 | Hurt |
| remove first and last resize in weight predictor (ablation) | Ours(3,10) 4K · HDR+ ablation | PSNR: 22.56 -> 22.65 | Improved |
| add color difference loss to basic loss (ablation) | Ours(3,10) · HDR+ (ours) | PSNR: 22.54 -> 22.61 | Improved |
| add perceptual loss to basic loss (ablation) | Ours(3,10) · HDR+ (ours) | LPIPS: 0.1906 -> 0.1470 | Improved |
| add both color and perceptual losses to basic loss (ablation) | Ours(3,10) · HDR+ (ours) | PSNR: 22.54 -> 22.73 | 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.
- DeepLPF — DeepLPF: Deep Local Parametric Filters for Image Enhancement (CVPR 2020) · better on 2 of 2 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
- Learning Image-adaptive 3D Lookup Tables for High Performance Photo Enhancement in Real-time (TPAMI 2020 · similarity 0.89)
- Real-Time Exposure Correction via Collaborative Transformations and Adaptive Sampling (CVPR 2024 · similarity 0.86)
- NILUT: Conditional Neural Implicit 3D Lookup Tables for Image Enhancement (AAAI 2024 · similarity 0.86)
- GLUT: 3D Gaussian Lookup Table for Continuous Color Transformation (NeurIPS 2026 · similarity 0.83)
- Reference-Based 3D-Aware Image Editing with Triplanes (CVPR 2025 · similarity 0.83)
- Ultra-High-Definition Dynamic Multi-Exposure Image Fusion via Infinite Pixel Learning (AAAI 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.
