Research map / Image generation / Editing and style transfer
Person Image Synthesis via Denoising Diffusion Model
Read the paperCheck 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 |
|---|---|---|---|
| Add a separate texture encoder and concatenate its features into the noise prediction layers (ablation) | PIDM baseline · DeepFashion | FID: 10.813 -> 9.8510 | Helped |
| Add texture diffusion blocks (ablation) | PIDM baseline with texture encoder · DeepFashion | FID: 9.8510 -> 7.5133 | Helped |
| Add vanilla classifier-free guidance (ablation) | PIDM with texture diffusion blocks · DeepFashion | FID: 7.5133 -> 6.8176 | Helped |
| Replace vanilla classifier-free guidance with disentangled classifier-free guidance (ablation) | PIDM with texture diffusion blocks · DeepFashion | FID: 6.8176 -> 6.3671 | Helped |
| Replace vanilla classifier-free guidance with disentangled classifier-free guidance (ablation) | PIDM with texture diffusion blocks · DeepFashion | SSIM: 0.7195 -> 0.7312 | Helped |
| Replace vanilla classifier-free guidance with disentangled classifier-free guidance (ablation) | PIDM with texture diffusion blocks · DeepFashion | LPIPS: 0.1769 -> 0.1676 | Helped |
| Add texture diffusion blocks (ablation) | PIDM baseline with texture encoder · DeepFashion | SSIM: 0.7005 -> 0.7178 | Helped |
| Add texture diffusion blocks (ablation) | PIDM baseline with texture encoder · DeepFashion | LPIPS: 0.1983 -> 0.1870 | Helped |
Closest papers by meaning
- Advancing Pose-Guided Image Synthesis with Progressive Conditional Diffusion Models (ICLR 2024 · similarity 0.92)
- Cross-view Masked Diffusion Transformers for Person Image Synthesis (ICML 2024 · similarity 0.91)
- Coarse-to-Fine Latent Diffusion for Pose-Guided Person Image Synthesis (CVPR 2024 · similarity 0.90)
- Multi-focal Conditioned Latent Diffusion for Person Image Synthesis (CVPR 2025 · similarity 0.90)
- IMAGPose: A Unified Conditional Framework for Pose-Guided Person Generation (NeurIPS 2024 · similarity 0.88)
- High-fidelity Person-centric Subject-to-Image Synthesis (CVPR 2024 · similarity 0.88)
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.