Research map / Image generation / Image restoration
StarEnhancer: Learning Real-Time and Style-Aware Image Enhancement
Yuda Song, Hui Qian, Xin Du
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 Dual AdaIN multi-style capability to the basic enhancer (ablation) | basic enhancer · MIT-Adobe-5K-UPE (single style, Artist C) | PSNR: 25.46 -> 25.29 | Hurt |
| add Dual AdaIN multi-style capability to the basic enhancer (ablation) | basic enhancer · MIT-Adobe-5K-UPE (single style, Artist C) | SSIM: 0.948 -> 0.943 | Hurt |
| add Dual AdaIN multi-style capability to the basic enhancer (ablation) | basic enhancer · MIT-Adobe-5K-UPE (single style, Artist C) | LPIPS: 0.083 -> 0.086 | Hurt |
| use StarEnhancer instead of Adaptive 3DLUT | StarEnhancer · MIT-Adobe-5K-UPE (single style, Artist C) | PSNR: 24.92 -> 25.29 | Improved |
| use StarEnhancer instead of Adaptive 3DLUT, speed | StarEnhancer · MIT-Adobe-5K-UPE (single style, Artist C), 4K images, TITAN RTX | FPS: 602 -> 205 | Hurt |
| use StarEnhancer instead of curve-based CURL | StarEnhancer · MIT-Adobe-5K-UPE (single style, Artist C) | PSNR: 24.20 -> 25.29 | Improved |
| use more sample images to compute latent code of unseen styles | StarEnhancer · MIT-Adobe-5K unseen styles | PSNR | Improved |
| unseen source style instead of unseen target style | StarEnhancer · MIT-Adobe-5K unseen styles | PSNR | Hurt |
Later papers that compared with this method
2 later accepted papers report StarEnhancer as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- PerTouch: VLM-Driven Agent for Personalized and Semantic Image Retouching (AAAI 2026 · better on 5 of 5 shared results)
- DiffRetouch: Using Diffusion to Retouch on the Shoulder of Experts (AAAI 2025 · better on 4 of 4 shared results)
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 3 of 3 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
- DeepLPF: Deep Local Parametric Filters for Image Enhancement (CVPR 2020 · similarity 0.86)
- Aesthetics-Guided Low-Light Enhancement (TPAMI 2025 · similarity 0.85)
- High-Resolution Photo Enhancement in Real-Time: A Laplacian Pyramid Network (TPAMI 2025 · similarity 0.84)
- Semantically Contrastive Learning for Low-Light Image Enhancement (AAAI 2022 · similarity 0.84)
- Attention Guided Low-Light Image Enhancement with a Large Scale Low-Light Simulation Dataset (IJCV 2021 · similarity 0.84)
- The Devil is in the Details: StyleFeatureEditor for Detail-Rich StyleGAN Inversion and High Quality Image Editing (CVPR 2024 · 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.
