Research map / 3D vision / Depth and stereo
Learning to Estimate Single-View Volumetric Flow Motions without 3D Supervision
Aleksandra Franz, Barbara Solenthaler, Nils Thuerey
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 |
|---|---|---|---|
| Replace linear interpolation with quadratic B-spline interpolation when upsampling velocity potentials (ablation) | Proposed method · ScalarFlow random view | FID: 117 -> 102 | Improved |
| Add the discriminator when no additional image targets are used (ablation) | Density reconstruction · Density reconstruction ablation | FID: 98 -> 68 | Improved |
| Remove additional image targets while retaining the discriminator (ablation) | Density reconstruction · Density reconstruction ablation | LPIPS: 0.124 -> 0.114 | Improved |
| Remove additional image targets while retaining the discriminator (ablation) | Density reconstruction · Density reconstruction ablation | FID: 80 -> 68 | Improved |
| Add additional image targets while retaining the discriminator (ablation) | Density reconstruction · Density reconstruction ablation | density RMSE | Improved |
Closest papers by meaning
- Volumetric Flow Estimation for Incompressible Fluids Using the Stationary Stokes Equations (ICCV 2017 · similarity 0.88)
- Learning Optical Flow From Still Images (CVPR 2021 · similarity 0.88)
- Global Transport for Fluid Reconstruction With Learned Self-Supervision (CVPR 2021 · similarity 0.88)
- Flow-Anything: Learning Real-World Optical Flow Estimation From Large-Scale Single-View Images (TPAMI 2025 · similarity 0.87)
- Self-Supervised Monocular Scene Flow Estimation (CVPR 2020 · similarity 0.86)
- Zero-Shot Monocular Scene Flow Estimation in the Wild (CVPR 2025 · similarity 0.84)
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.
