FlexPainter: Flexible and Multi-View Consistent Texture Generation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915320965365760 |
|---|---|
| author | Yan, Dongyu Wu, Leyi Lin, Jiantao Wang, Luozhou Xu, Tianshuo Chen, Zhifei Yang, Zhen Xu, Lie Zhang, Shunsi Chen, Yingcong |
| author_facet | Yan, Dongyu Wu, Leyi Lin, Jiantao Wang, Luozhou Xu, Tianshuo Chen, Zhifei Yang, Zhen Xu, Lie Zhang, Shunsi Chen, Yingcong |
| contents | Texture map production is an important part of 3D modeling and determines the rendering quality. Recently, diffusion-based methods have opened a new way for texture generation. However, restricted control flexibility and limited prompt modalities may prevent creators from producing desired results. Furthermore, inconsistencies between generated multi-view images often lead to poor texture generation quality. To address these issues, we introduce \textbf{FlexPainter}, a novel texture generation pipeline that enables flexible multi-modal conditional guidance and achieves highly consistent texture generation. A shared conditional embedding space is constructed to perform flexible aggregation between different input modalities. Utilizing such embedding space, we present an image-based CFG method to decompose structural and style information, achieving reference image-based stylization. Leveraging the 3D knowledge within the image diffusion prior, we first generate multi-view images simultaneously using a grid representation to enhance global understanding. Meanwhile, we propose a view synchronization and adaptive weighting module during diffusion sampling to further ensure local consistency. Finally, a 3D-aware texture completion model combined with a texture enhancement model is used to generate seamless, high-resolution texture maps. Comprehensive experiments demonstrate that our framework significantly outperforms state-of-the-art methods in both flexibility and generation quality. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_02620 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | FlexPainter: Flexible and Multi-View Consistent Texture Generation Yan, Dongyu Wu, Leyi Lin, Jiantao Wang, Luozhou Xu, Tianshuo Chen, Zhifei Yang, Zhen Xu, Lie Zhang, Shunsi Chen, Yingcong Graphics Computer Vision and Pattern Recognition Texture map production is an important part of 3D modeling and determines the rendering quality. Recently, diffusion-based methods have opened a new way for texture generation. However, restricted control flexibility and limited prompt modalities may prevent creators from producing desired results. Furthermore, inconsistencies between generated multi-view images often lead to poor texture generation quality. To address these issues, we introduce \textbf{FlexPainter}, a novel texture generation pipeline that enables flexible multi-modal conditional guidance and achieves highly consistent texture generation. A shared conditional embedding space is constructed to perform flexible aggregation between different input modalities. Utilizing such embedding space, we present an image-based CFG method to decompose structural and style information, achieving reference image-based stylization. Leveraging the 3D knowledge within the image diffusion prior, we first generate multi-view images simultaneously using a grid representation to enhance global understanding. Meanwhile, we propose a view synchronization and adaptive weighting module during diffusion sampling to further ensure local consistency. Finally, a 3D-aware texture completion model combined with a texture enhancement model is used to generate seamless, high-resolution texture maps. Comprehensive experiments demonstrate that our framework significantly outperforms state-of-the-art methods in both flexibility and generation quality. |
| title | FlexPainter: Flexible and Multi-View Consistent Texture Generation |
| topic | Graphics Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2506.02620 |