Research map / 3D vision / Face and shape
FluidGaussian: Propagating Simulation-Based Uncertainty Toward Functionally-Intelligent 3D Reconstruction
Yuqiu Liu, Jialin Song, Marissa Ramirez de Chanlatte, Rochishnu Chowdhury, Rushil Paresh Desai, Wuyang Chen, Daniel Martin, Michael W. Mahoney
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 FluidGaussian view selection | FisherRF · Blender | PSNR: 23.4230 -> 24.7433 | Improved |
| add FluidGaussian view selection | ActiveNeRF · DrivAerNet++ | PSNR: 17.3840 -> 18.8727 | Improved |
| add FluidGaussian view selection | ActiveNeRF · MipNeRF360 | PSNR: 12.9245 -> 12.4790 | Hurt |
| add FluidGaussian view selection | FisherRF · MipNeRF360 | PSNR: 15.3237 -> 15.5502 | No clear effect |
| add FluidGaussian view selection | ActiveNeRF · Blender | mean divergence: 0.0262 -> 0.0145 | Improved |
| add FluidGaussian view selection | ActiveNeRF · MipNeRF360 | mean divergence: 0.0130 -> 0.0049 | Improved |
| add FluidGaussian view selection | FisherRF · DrivAerNet++ | mean divergence: 0.0609 -> 0.0608 | No clear effect |
| add FluidGaussian view selection | FisherRF · DrivAerNet++ function-critical views | PSNR: 18.2988 -> 18.5436 | Improved |
| add FluidGaussian view selection | ActiveNeRF · DrivAerNet++ function-critical views | divergence: 0.0316 -> 0.0231 | Improved |
| use all five fluid directions instead of one (ablation) | FisherRF + FluidGaussian · Blender | PSNR: 24.0015-24.5565 (single) -> 24.7433 | Improved |
| use single fluid direction instead of no physics (ablation) | FisherRF · Blender | PSNR: 23.4230 -> 24.0015 to 24.5565 | Improved |
| add FluidGaussian at higher Reynolds numbers | ActiveNeRF · Blender | divergence gap | 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.
- ActiveNeRF — ActiveNeRF: Learning Where to See with Uncertainty Estimation (ECCV 2022) · better on 8 of 8 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
- FluidNexus: 3D Fluid Reconstruction and Prediction from a Single Video (CVPR 2025 · similarity 0.88)
- DecoupledGaussian: Object-Scene Decoupling for Physics-Based Interaction (CVPR 2025 · similarity 0.87)
- PhysGaussian: Physics-Integrated 3D Gaussians for Generative Dynamics (CVPR 2024 · similarity 0.87)
- OmniPhysGS: 3D Constitutive Gaussians for General Physics-Based Dynamics Generation (ICLR 2025 · similarity 0.86)
- PhyRecon: Physically Plausible Neural Scene Reconstruction (NeurIPS 2024 · similarity 0.86)
- MAtCha Gaussians: Atlas of Charts for High-Quality Geometry and Photorealism From Sparse Views (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.
