Research map / Image generation / Editing and style transfer
MUST-GAN: Multi-Level Statistics Transfer for Self-Driven Person Image Generation
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 multi-level statistics transfer instead of transferring statistics only from the final appearance-encoder level (ablation) | MUST-GAN · DeepFashion pose transfer | FID: 21.928 -> 15.902 | Helped |
| Add channel attention (ablation) | MUST-GAN · DeepFashion pose transfer | SSIM: 0.737 -> 0.742 | Helped |
| Add channel attention (ablation) | MUST-GAN · DeepFashion pose transfer | IS: 3.729 -> 3.692 | Hurt |
| Add pose connection map (ablation) | MUST-GAN · DeepFashion pose transfer | SSIM: 0.724 -> 0.742 | Helped |
| Add pose connection map (ablation) | MUST-GAN · DeepFashion pose transfer | IS: 3.649 -> 3.692 | Helped |
| Train self-driven without paired source-target images or target-pose information | MUST-GAN · DeepFashion pose transfer | IS: 3.692 vs. 3.441 | Helped |
| Use self-driven training rather than paired supervised training | MUST-GAN · DeepFashion pose transfer | FID: 15.902 vs. 14.061 | Mixed |
Closest papers by meaning
- Controllable Person Image Synthesis With Attribute-Decomposed GAN (CVPR 2020 · similarity 0.88)
- Pose Guided Person Image Generation (NeurIPS 2017 · similarity 0.88)
- Progressive Pose Attention Transfer for Person Image Generation (CVPR 2019 · similarity 0.88)
- Human Appearance Transfer (CVPR 2018 · similarity 0.88)
- Disentangled Person Image Generation (CVPR 2018 · similarity 0.87)
- IMAGPose: A Unified Conditional Framework for Pose-Guided Person Generation (NeurIPS 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.