Research map / Label-efficient and robust learning / Self-supervised and contrastive
Enlightening Photographic Style Transfer with a Self-Supervised Photographic Embedding
Chengxuan Zhu, Jiacong Fang, Shuchen Weng, Youwei Lyu, Jiajun Tang, Qingnan Fan, Chao Xu, Boxin Shi
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 |
|---|---|---|---|
| use CIE-Lab instead of RGB color space (ablation) | PETAL embedding · PPR10K subset retrieval | R@1: 70.34 -> 74.27 | Improved |
| use CIE-Lab instead of RGB color space (ablation) | PETAL · PST50 style transfer | PSNR: 22.22 -> 24.63 | Improved |
| add cross-attention histogram injection (ablation) | PETAL embedding · PPR10K subset retrieval | R@1: 68.93 -> 74.27 | Improved |
| add cross-attention histogram injection (ablation) | PETAL · PST50 style transfer | PSNR: 22.24 -> 24.63 | Improved |
| use adaptive pooling instead of mean pooling (ablation) | PETAL embedding · PPR10K subset retrieval | R@1: 68.91 -> 74.27 | Improved |
| add pooled patch features to the CLS token (ablation) | PETAL embedding · PPR10K subset retrieval | R@1: 73.74 -> 74.27 | Improved |
| add pooled patch features to the CLS token (ablation) | PETAL · PST50 style transfer | PSNR: 23.11 -> 24.63 | Improved |
| add histogram reconstruction loss (ablation) | PETAL embedding · PPR10K subset retrieval | R@1: 66.98 -> 74.27 | Improved |
| add histogram reconstruction loss (ablation) | PETAL · PST50 style transfer | LPIPS: 0.1354 -> 0.1021 | Improved |
| add intra-sample loss (ablation) | PETAL embedding · PPR10K subset retrieval | R@1: 36.09 -> 74.27 | Improved |
| add inter-sample loss with hard negatives (ablation) | PETAL embedding · PPR10K subset retrieval | R@1: 67.53 -> 74.27 | Improved |
| fine-tune DINOv3 with the same data and objectives | DINOv3 · PPR10K subset retrieval | R@1: 1.48 -> 59.30 | Improved |
Closest papers by meaning
- Towards In-Context Tone Style Transfer with A Large-Scale Triplet Dataset (ECCV 2026 · similarity 0.88)
- Deep Photo Style Transfer (CVPR 2017 · similarity 0.88)
- Artistic Style Transfer with Internal-external Learning and Contrastive Learning (NeurIPS 2021 · similarity 0.88)
- Automatic Content-Aware Color and Tone Stylization (CVPR 2016 · similarity 0.88)
- SigStyle: Signature Style Transfer via Personalized Text-to-Image Models (AAAI 2025 · similarity 0.86)
- Style Injection in Diffusion: A Training-free Approach for Adapting Large-scale Diffusion Models for Style Transfer (CVPR 2024 · similarity 0.86)
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.
