EpiDiff: Enhancing Multi-View Synthesis via Localized Epipolar-Constrained Diffusion
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2023
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866913294600634368 |
|---|---|
| author | Huang, Zehuan Wen, Hao Dong, Junting Wang, Yaohui Li, Yangguang Chen, Xinyuan Cao, Yan-Pei Liang, Ding Qiao, Yu Dai, Bo Sheng, Lu |
| author_facet | Huang, Zehuan Wen, Hao Dong, Junting Wang, Yaohui Li, Yangguang Chen, Xinyuan Cao, Yan-Pei Liang, Ding Qiao, Yu Dai, Bo Sheng, Lu |
| contents | Generating multiview images from a single view facilitates the rapid generation of a 3D mesh conditioned on a single image. Recent methods that introduce 3D global representation into diffusion models have shown the potential to generate consistent multiviews, but they have reduced generation speed and face challenges in maintaining generalizability and quality. To address this issue, we propose EpiDiff, a localized interactive multiview diffusion model. At the core of the proposed approach is to insert a lightweight epipolar attention block into the frozen diffusion model, leveraging epipolar constraints to enable cross-view interaction among feature maps of neighboring views. The newly initialized 3D modeling module preserves the original feature distribution of the diffusion model, exhibiting compatibility with a variety of base diffusion models. Experiments show that EpiDiff generates 16 multiview images in just 12 seconds, and it surpasses previous methods in quality evaluation metrics, including PSNR, SSIM and LPIPS. Additionally, EpiDiff can generate a more diverse distribution of views, improving the reconstruction quality from generated multiviews. Please see our project page at https://huanngzh.github.io/EpiDiff/. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2312_06725 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | EpiDiff: Enhancing Multi-View Synthesis via Localized Epipolar-Constrained Diffusion Huang, Zehuan Wen, Hao Dong, Junting Wang, Yaohui Li, Yangguang Chen, Xinyuan Cao, Yan-Pei Liang, Ding Qiao, Yu Dai, Bo Sheng, Lu Computer Vision and Pattern Recognition Generating multiview images from a single view facilitates the rapid generation of a 3D mesh conditioned on a single image. Recent methods that introduce 3D global representation into diffusion models have shown the potential to generate consistent multiviews, but they have reduced generation speed and face challenges in maintaining generalizability and quality. To address this issue, we propose EpiDiff, a localized interactive multiview diffusion model. At the core of the proposed approach is to insert a lightweight epipolar attention block into the frozen diffusion model, leveraging epipolar constraints to enable cross-view interaction among feature maps of neighboring views. The newly initialized 3D modeling module preserves the original feature distribution of the diffusion model, exhibiting compatibility with a variety of base diffusion models. Experiments show that EpiDiff generates 16 multiview images in just 12 seconds, and it surpasses previous methods in quality evaluation metrics, including PSNR, SSIM and LPIPS. Additionally, EpiDiff can generate a more diverse distribution of views, improving the reconstruction quality from generated multiviews. Please see our project page at https://huanngzh.github.io/EpiDiff/. |
| title | EpiDiff: Enhancing Multi-View Synthesis via Localized Epipolar-Constrained Diffusion |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2312.06725 |