Research map / Image generation / Editing and style transfer
Toward Tiny and High-quality Facial Makeup with Data Amplify Learning
Qiaoqiao Jin, Xuanhong Chen, Meiguang Jin, Ying Chen, Rui Shi, Yucheng Zheng, Yupeng Zhu, Bingbing Ni
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 EleGANt with DAL-amplified paired data instead of pseudo-paired data | EleGANt · FFHQ test | PSNR: 30.18 -> 35.45 | Improved |
| train BeautyREC with DAL-amplified paired data | BeautyREC · FFHQ test | PSNR: 24.93 -> 31.06 | Improved |
| train BeautyREC with DAL data | BeautyREC · MT test | PSNR: 27.39 -> 27.39 | No clear effect |
| replace EleGANt with the tiny 14-conv TinyBeauty trained on DAL data | TinyBeauty · FFHQ test | PSNR: 30.18 -> 35.39 | Improved |
| TinyBeauty vs EleGANt trained on same DAL data | TinyBeauty · FFHQ test | PSNR: 35.45 -> 35.39 | No clear effect |
| add eyeliner edge (Sobel) loss to TinyBeauty (ablation) | TinyBeauty · visual eyeliner clarity | qualitative eyeliner clarity | Improved |
| combine mask guidance, IPB, SPB and RDM in the data amplifier (ablation) | DDA · qualitative DDA ablation | identity, style and texture fidelity | Improved |
| TinyBeauty uses 14 conv layers and no face prompts versus prior multi-model pipelines | TinyBeauty · iPhone13 latency | RunTime (ms): 27.89 -> 2.18 | 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.
- EleGANt — ELEGANT: Exchanging Latent Encodings with GAN for Transferring Multiple Face Attributes (ECCV 2018) · better on 6 of 6 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
- SHMT: Self-supervised Hierarchical Makeup Transfer via Latent Diffusion Models (NeurIPS 2024 · similarity 0.88)
- Makeup Prior Models for 3D Facial Makeup Estimation and Applications (CVPR 2024 · similarity 0.85)
- DiffAM: Diffusion-based Adversarial Makeup Transfer for Facial Privacy Protection (CVPR 2024 · similarity 0.84)
- FSFM: A Generalizable Face Security Foundation Model via Self-Supervised Facial Representation Learning (CVPR 2025 · similarity 0.84)
- OSDFace: One-Step Diffusion Model for Face Restoration (CVPR 2025 · similarity 0.84)
- Data Synthesis with Diverse Styles for Face Recognition via 3DMM-Guided Diffusion (CVPR 2025 · similarity 0.84)
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.
