Research map / 3D vision / Depth and stereo
3D Fluid Flow Estimation with Integrated Particle Reconstruction
Katrin Lasinger, Christoph Vogel, Thomas Pock, Konrad Schindler B
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 hard divergence constraint (vs none) (ablation) | IPR flow estimation · JHTDB 0.1 ppp | AEE: 0.157 -> 0.136 | Improved |
| use hard divergence instead of soft divergence alpha=64 (ablation) | IPR flow estimation · JHTDB 0.1 ppp | AEE: 0.136 vs 0.135 | No clear effect |
| use joint particle and flow model instead of previous sequential Eulerian approach | IPR joint · JHTDB 0.1 ppp | AEE: 0.406 -> 0.136 | Improved |
| optimize particles and flow jointly instead of sequentially (ablation) | IPR joint · JHTDB 0.2 ppp | AEE: 0.407 -> 0.108 | Improved |
| use iterative triangulation instead of MART initialization (ablation) | IPR joint · JHTDB 0.2 ppp | AEE: 0.506 -> 0.108 | Improved |
| use 0-norm instead of 1-norm sparsity (ablation) | IPR joint · JHTDB 0.2 ppp | particle precision (%): 25.23 -> 55.43 | Improved |
| use 0-norm instead of 1-norm sparsity (ablation) | IPR joint · JHTDB 0.2 ppp | AEE: 0.110 -> 0.108 | Improved |
| increase particle density from 0.1 to 0.2 ppp | IPR joint · JHTDB simulated | AEE: 0.136 -> 0.108 | Improved |
| increase particle density from 0.2 to 0.25 ppp | IPR joint · JHTDB simulated | AEE: 0.108 -> 0.134 | Hurt |
| increase particle size sigma from 1 to 1.6 | IPR joint · JHTDB 0.1 ppp | AEE: 0.136 -> 0.235 | Hurt |
| decrease particle size sigma from 1 to 0.8 | IPR joint · JHTDB 0.1 ppp | AEE: 0.136 -> 0.135 | No clear effect |
| use unequal particle sizes N(1,0.1^2) with fixed sigma=1 at inference | IPR joint · JHTDB 0.1 ppp | AEE: 0.136 -> 0.155 | Hurt |
Closest papers by meaning
- Volumetric Flow Estimation for Incompressible Fluids Using the Stationary Stokes Equations (ICCV 2017 · similarity 0.91)
- Full-Volume 3D Fluid Flow Reconstruction With Light Field PIV (TPAMI 2023 · similarity 0.91)
- 3D Fluid Flow Reconstruction Using Compact Light Field PIV (ECCV 2020 · similarity 0.89)
- Dual-frame Fluid Motion Estimation with Test-time Optimization and Zero-divergence Loss (NeurIPS 2024 · similarity 0.87)
- FluidNexus: 3D Fluid Reconstruction and Prediction from a Single Video (CVPR 2025 · similarity 0.85)
- DeepLag: Discovering Deep Lagrangian Dynamics for Intuitive Fluid Prediction (NeurIPS 2024 · similarity 0.85)
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.
