Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2411.17189 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915038916247552 |
|---|---|
| author | Tan, Xiyang Jiang, Ying Li, Xuan Zong, Zeshun Xie, Tianyi Yang, Yin Jiang, Chenfanfu |
| author_facet | Tan, Xiyang Jiang, Ying Li, Xuan Zong, Zeshun Xie, Tianyi Yang, Yin Jiang, Chenfanfu |
| contents | We introduce PhysMotion, a novel framework that leverages principled physics-based simulations to guide intermediate 3D representations generated from a single image and input conditions (e.g., applied force and torque), producing high-quality, physically plausible video generation. By utilizing continuum mechanics-based simulations as a prior knowledge, our approach addresses the limitations of traditional data-driven generative models and result in more consistent physically plausible motions. Our framework begins by reconstructing a feed-forward 3D Gaussian from a single image through geometry optimization. This representation is then time-stepped using a differentiable Material Point Method (MPM) with continuum mechanics-based elastoplasticity models, which provides a strong foundation for realistic dynamics, albeit at a coarse level of detail. To enhance the geometry, appearance and ensure spatiotemporal consistency, we refine the initial simulation using a text-to-image (T2I) diffusion model with cross-frame attention, resulting in a physically plausible video that retains intricate details comparable to the input image. We conduct comprehensive qualitative and quantitative evaluations to validate the efficacy of our method. Our project page is available at: https://supertan0204.github.io/physmotion_website/. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_17189 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | PhysMotion: Physics-Grounded Dynamics From a Single Image Tan, Xiyang Jiang, Ying Li, Xuan Zong, Zeshun Xie, Tianyi Yang, Yin Jiang, Chenfanfu Computer Vision and Pattern Recognition We introduce PhysMotion, a novel framework that leverages principled physics-based simulations to guide intermediate 3D representations generated from a single image and input conditions (e.g., applied force and torque), producing high-quality, physically plausible video generation. By utilizing continuum mechanics-based simulations as a prior knowledge, our approach addresses the limitations of traditional data-driven generative models and result in more consistent physically plausible motions. Our framework begins by reconstructing a feed-forward 3D Gaussian from a single image through geometry optimization. This representation is then time-stepped using a differentiable Material Point Method (MPM) with continuum mechanics-based elastoplasticity models, which provides a strong foundation for realistic dynamics, albeit at a coarse level of detail. To enhance the geometry, appearance and ensure spatiotemporal consistency, we refine the initial simulation using a text-to-image (T2I) diffusion model with cross-frame attention, resulting in a physically plausible video that retains intricate details comparable to the input image. We conduct comprehensive qualitative and quantitative evaluations to validate the efficacy of our method. Our project page is available at: https://supertan0204.github.io/physmotion_website/. |
| title | PhysMotion: Physics-Grounded Dynamics From a Single Image |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2411.17189 |