Research map / 3D vision / Face and shape
A Riemannian Framework for the Elastic Analysis of the Spatiotemporal Variability in the Shape and Structure of Tree-Like 4D Objects
Tahmina Khanam, Hamid Laga, Mohammed Bennamoun, Guanjin Wang, Ferdous Ahmed Sohel, Farid Boussaïd, Guan Wang, Anuj Srivastava
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 Pan et al. with ESRVF-based spatial registration | spatial registration · Globe plants (eps=0.1) | cycle consistency error, mean: 20% -> 0.08% | Improved |
| replace Pan et al. with ESRVF-based spatial registration | spatial registration · Pheno4D (eps=0.01) | cycle consistency error, mean: 21.95% -> 5.45% | Improved |
| replace Pan et al. with ESRVF-based spatial registration | spatial registration · Pheno4D | time per 3D tree: 20.1 s -> 2.3 s | Improved |
| replace Pan et al. with ESRVF-based spatial registration | spatial registration · Globe plants | time per 3D tree: 3 minutes to 6 hours -> 5 s to 2.7 hours | Improved |
| add temporal registration between source and target 4D trees | SRVF temporal registration · Globe plants | geodesic distance, mean: 159 -> 67.7 | Improved |
| add temporal registration between source and target 4D trees | SRVF temporal registration · Pheno4D Tomato | geodesic distance, mean: 382 -> 279 | Improved |
| add temporal registration to randomly warped 4D tree | SRVF temporal registration · Pheno4D Tomato | geodesic distance, mean: 356 -> 26 | Improved |
| add temporal registration to randomly warped 4D tree | SRVF temporal registration · Globe plants | geodesic distance, mean: 77.8 -> 14.8 | Improved |
| do temporal registration in SRVF space instead of original PCA curve space (ablation) | temporal registration · warped 4D tree | synchronization | Improved |
| add Yeo-Johnson power transformation before PCA (ablation) | PCA subspace of 3D trees · 4D tree reconstruction | reconstruction quality | Improved |
| use L2 metric in ESRVF space instead of original tree space for interpolation (ablation) | tree geodesics · 3D tree interpolation | shrinkage | Improved |
| perform temporal registration in PCA space instead of ESRVFT space | temporal registration · 4D tree pairs | computation time: 10 to 20 minutes -> 4 to 30 ms | Improved |
Closest papers by meaning
- A Riemannian Approach for Spatiotemporal Analysis and Generation of 4D Tree-shaped Structures (ECCV 2024 · similarity 0.96)
- Elastic Shape Analysis of Tree-Like 3D Objects Using Extended SRVF Representation (TPAMI 2023 · similarity 0.93)
- TreeSRNF: Square-Root Normal Fields for Generative Modelling of the Geometric and Structural Variability in Tree-like 3D Objects (ECCV 2026 · similarity 0.93)
- 4D Atlas: Statistical Analysis of the Spatiotemporal Variability in Longitudinal 3D Shape Data (TPAMI 2022 · similarity 0.91)
- Dynamic Neural Surfaces for Elastic 4D Shape Representation and Analysis (CVPR 2025 · similarity 0.87)
- Geometry-Centered 3D Latent World Models for Growing Surfaces (NeurIPS 2026 · similarity 0.83)
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.
