Research map / 3D vision / 3D generation
Learning an Implicit Physics Model for Image-based Fluid Simulation
Emily Yue-Ting Jia, Jiageng Mao, Zhiyuan Gao, Yajie Zhao, Yue Wang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 3D Gaussians (without physics) (ablation) | image-to-4D fluid animation · Holynski val (ablation) | PSNR: 21.49 -> 22.58 | Improved |
| add physics-informed neural dynamics on top of 3D Gaussians (ablation) | image-to-4D fluid animation · Holynski val (ablation) | PSNR: 22.58 -> 23.05 | Improved |
| add physics dynamics on top of Gaussians (ablation) | image-to-4D fluid animation · Holynski val (ablation) | LPIPS: 0.22 -> 0.26 | Hurt |
| add viscosity to physics guidance (ablation) | physics-informed dynamics · Synthetic scenes w/ rock | velocity L1 error: 0.0388 -> 0.0501 | Hurt |
| add pressure fields to physics guidance (ablation) | physics-informed dynamics · Synthetic scenes w/ rock | velocity L1 error: 0.0388 -> 0.415 | Hurt |
| add external forces (gravity) to physics guidance (ablation) | physics-informed dynamics · Synthetic scenes w/o rock | velocity L1 error: 0.0612 -> 0.0389 | Improved |
| use physics-informed dynamics instead of 3D-Cinemagraphy motion | fluid velocity prediction · Synthetic scenes w/ rock | velocity L1 error: 0.0498 -> 0.0388 | Improved |
| use ours instead of 3D-Cinemagraphy | image-to-video fluid animation · Holynski val input view | PSNR: 22.81 -> 24.98 | Improved |
| use ours instead of 3D-Cinemagraphy | image-to-video fluid animation · Holynski val novel view | PSNR: 22.46 -> 24.34 | Improved |
Closest papers by meaning
- Latent Intuitive Physics: Learning to Transfer Hidden Physics from A 3D Video (ICLR 2024 · similarity 0.89)
- DreamPhysics: Learning Physics-Based 3D Dynamics with Video Diffusion Priors (AAAI 2025 · similarity 0.88)
- FreeGave: 3D Physics Learning from Dynamic Videos by Gaussian Velocity (CVPR 2025 · similarity 0.88)
- Learning 3D Particle-based Simulators from RGB-D Videos (ICLR 2024 · similarity 0.87)
- Real-time Photorealistic Dynamic Scene Representation and Rendering with 4D Gaussian Splatting (ICLR 2024 · similarity 0.86)
- FluidNexus: 3D Fluid Reconstruction and Prediction from a Single Video (CVPR 2025 · similarity 0.86)
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.
