Research map / 3D vision / Gaussian splatting
GaussianFluent: Gaussian Simulation for Dynamic Scenes with Mixed Materials
Bei Huang, Yixin Chen, Ruijie Lu, Gang Zeng, Hongbin Zha, Yuru Pei, Siyuan Huang
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 |
|---|---|---|---|
| synthesize internal textures via multi-axis inpainting instead of copying exterior colors | GaussianFluent · Internal texture filling | CLIP score: 22.3 -> 35.4 | Improved |
| use iterative 3D-consistent inpainting instead of 2D slice inpainting | GaussianFluent · Internal texture filling | CLIP score: 30.1 -> 35.4 | Improved |
| use brittle-fracture CD-MPM vs PhysGaussian | GaussianFluent · Dynamic scene simulation | CLIP score: 12.2 -> 22.7 | Improved |
| use GaussianFluent vs OmniPhysGS | GaussianFluent · Dynamic scene simulation | user preference (%): 7.69% -> 88.46% | Improved |
| assign per-part damage parameter (rind, flesh, seed) instead of uniform epsilon (ablation) | GaussianFluent · watermelon fracture | visual realism (qualitative) | Improved |
| replace fixed return-mapping center with dynamic point (k=2) (ablation) | CD-MPM for GS · simulation stability | stability | Improved |
| run CD-MPM on GPU with continuous return mapping instead of CPU | GaussianFluent · MPM grid 300^3 | simulation time (s/frame): 3408 -> 5.05 | Improved |
| add damage field tracking over PhysGaussian | GaussianFluent · MPM grid 300^3 | simulation time (s/frame): 4.78 -> 5.05 | Hurt |
| add damage field tracking over PhysGaussian | GaussianFluent · MPM grid 200^3 | simulation time (s/frame): 2.12 -> 2.08 | No clear effect |
| add opacity-only optimization after internal filling (ablation) | GaussianFluent · internal volume | surface smoothness (qualitative) | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- OmniPhysGS — OmniPhysGS: 3D Constitutive Gaussians for General Physics-Based Dynamics Generation (ICLR 2025) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Gaussian Splashing: Unified Particles for Versatile Motion Synthesis and Rendering (CVPR 2025 · similarity 0.90)
- DisC-GS: Discontinuity-aware Gaussian Splatting (NeurIPS 2024 · similarity 0.89)
- Reflective Gaussian Splatting (ICLR 2025 · similarity 0.89)
- Fully Explicit Dynamic Gaussian Splatting (NeurIPS 2024 · similarity 0.89)
- Ref-GS: Directional Factorization for 2D Gaussian Splatting (CVPR 2025 · similarity 0.89)
- GaussianPro: 3D Gaussian Splatting with Progressive Propagation (ICML 2024 · similarity 0.88)
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.
