Research map / Image generation / Editing and style transfer
Progressive and Aligned Pose Attention Transfer for Person Image Generation
Zhen Zhu, Tengteng Huang, Mengde Xu, Baoguang Shi, Wenqing Cheng, Xiang Bai
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 |
|---|---|---|---|
| replace residual blocks with PATBs (9 blocks) (ablation) | Resnet generator · Market-1501 | SSIM: 0.301 -> 0.311 | Improved |
| replace residual blocks with PATBs (9 blocks) (ablation) | Resnet generator · DeepFashion | PCKh: 0.90 -> 0.96 | Improved |
| increase PATN blocks from 5 to 9 (ablation) | PATN · Market-1501 | SSIM: 0.309 -> 0.311 | Improved |
| increase PATN blocks from 9 to 13 (ablation) | PATN · Market-1501 | SSIM: 0.311 -> 0.314 | No clear effect |
| increase Resnet blocks from 9 to 13 (ablation) | Resnet generator · Market-1501 | SSIM: 0.301 -> 0.300 | No clear effect |
| use 5 APATBs instead of 1 (ablation) | APATN · Market-1501 | SSIM: 0.251 -> 0.326 | Improved |
| use 9 APATBs instead of 5 (ablation) | APATN · Market-1501 | SSIM: 0.326 -> 0.326 | No clear effect |
| add addition path in PATB (ablation) | PATB · Market-1501 | SSIM: 0.305 -> 0.311 | Improved |
| add concatenation path in PATB (ablation) | PATB · Market-1501 | SSIM: 0.306 -> 0.311 | Improved |
| add concatenation in APATB image code update (ablation) | APATB · Market-1501 | SSIM: 0.233 -> 0.326 | Improved |
| add residual blocks to update pose code in APATB (ablation) | APATB · Market-1501 | SSIM: 0.255 -> 0.326 | Improved |
| use concatenation instead of addition in APATB (ablation) | APATB · Market-1501 | SSIM: 0.333 -> 0.326 | Mixed |
Closest papers by meaning
- Progressive Pose Attention Transfer for Person Image Generation (CVPR 2019 · similarity 0.96)
- Pose Guided Person Image Generation (NeurIPS 2017 · similarity 0.89)
- MUST-GAN: Multi-Level Statistics Transfer for Self-Driven Person Image Generation (CVPR 2021 · similarity 0.88)
- Pose-Normalized Image Generation for Person Re-identification (ECCV 2018 · similarity 0.88)
- Appearance and Pose-Conditioned Human Image Generation Using Deformable GANs (TPAMI 2019 · similarity 0.88)
- 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.
