Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2507.13311 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915396505829376 |
|---|---|
| author | Shi, Chuancheng Chen, Yixiang Lei, Burong Chen, Jichao |
| author_facet | Shi, Chuancheng Chen, Yixiang Lei, Burong Chen, Jichao |
| contents | Realistic and controllable garment visualization is critical for fashion e-commerce, where users expect personalized previews under diverse poses and lighting conditions. Existing methods often rely on predefined poses, limiting semantic flexibility and illumination adaptability. To address this, we introduce FashionPose, the first unified text-to-pose-to-relighting generation framework. Given a natural language description, our method first predicts a 2D human pose, then employs a diffusion model to generate high-fidelity person images, and finally applies a lightweight relighting module, all guided by the same textual input. By replacing explicit pose annotations with text-driven conditioning, FashionPose enables accurate pose alignment, faithful garment rendering, and flexible lighting control. Experiments demonstrate fine-grained pose synthesis and efficient, consistent relighting, providing a practical solution for personalized virtual fashion display. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_13311 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | FashionPose: Text to Pose to Relight Image Generation for Personalized Fashion Visualization Shi, Chuancheng Chen, Yixiang Lei, Burong Chen, Jichao Computer Vision and Pattern Recognition Realistic and controllable garment visualization is critical for fashion e-commerce, where users expect personalized previews under diverse poses and lighting conditions. Existing methods often rely on predefined poses, limiting semantic flexibility and illumination adaptability. To address this, we introduce FashionPose, the first unified text-to-pose-to-relighting generation framework. Given a natural language description, our method first predicts a 2D human pose, then employs a diffusion model to generate high-fidelity person images, and finally applies a lightweight relighting module, all guided by the same textual input. By replacing explicit pose annotations with text-driven conditioning, FashionPose enables accurate pose alignment, faithful garment rendering, and flexible lighting control. Experiments demonstrate fine-grained pose synthesis and efficient, consistent relighting, providing a practical solution for personalized virtual fashion display. |
| title | FashionPose: Text to Pose to Relight Image Generation for Personalized Fashion Visualization |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2507.13311 |