Research map / 3D vision / Face and shape
A Kendall Shape Space Approach to 3D Shape Estimation from 2D Landmarks
Martha Paskin, Daniel Baum, Mason N. Dean, Christoph von Tycowicz
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 ASM with Kendall shape space barycentric model | ASM non-convex · MoCap human pose, subject 13, 32 basis shapes | mean Procrustes error: 0.462(0.121) -> 0.295(0.031) | Improved |
| replace linear ASM with Kendall shape space barycentric model | ASM convex · MoCap human pose, subject 14, 128 basis shapes | mean Procrustes error: 0.533(0.087) -> 0.242(0.019) | Improved |
| replace linear ASM with Kendall shape space model | ASM non-convex · MoCap human pose, subject 15, 128 basis shapes | mean Procrustes error: 0.457(0.089) -> 0.221(0.021) | Improved |
| increase number of basis shapes 32 to 128 (ablation) | ASM non-convex · MoCap human pose, subject 14 | mean Procrustes error: 0.475(0.129) -> 0.644(0.128) | Hurt |
| increase number of basis shapes 32 to 128 (ablation) | ASM convex · MoCap human pose, subject 13 | mean Procrustes error: 0.550(0.071) -> 0.461(0.059) | Improved |
| increase number of basis shapes 32 to 128 (ablation) | KSS · MoCap human pose, subject 13 | mean Procrustes error: 0.295(0.031) -> 0.288(0.038) | No clear effect |
| replace ASM non-convex with KSS | ASM non-convex · Basking shark BS1, 15 basis, symmetric | mean Procrustes error: 0.231(0.002) -> 0.215(0.001) | Improved |
| replace ASM non-convex with KSS | ASM non-convex · Basking shark BS1, 29 basis, asymmetric | mean Procrustes error: 0.136(2e-04) -> 0.156(0.0003) | Hurt |
| replace ASM non-convex with KSS | ASM non-convex · Basking shark BS2, 29 basis, symmetric | mean Procrustes error: 0.281(0.014) -> 0.256(0.012) | Improved |
| replace ASM convex with KSS | ASM convex · Basking shark BS1, 15 basis, symmetric | mean Procrustes error: 0.854(0.030) -> 0.215(0.001) | Improved |
| use KSS instead of linear ASM on real photographs | ASM approaches · qualitative real human and underwater shark images | 3D plausibility | Improved |
Closest papers by meaning
- Lions and Tigers and Bears: Capturing Non-Rigid, 3D, Articulated Shape From Images (CVPR 2018 · similarity 0.88)
- Animal3D: A Comprehensive Dataset of 3D Animal Pose and Shape (ICCV 2023 · similarity 0.88)
- Model-based Metric 3D Shape and Motion Reconstruction of Wild Bottlenose Dolphins in Drone-Shot Videos (IJCV 2026 · similarity 0.87)
- Birds of a Feather: Capturing Avian Shape Models From Images (CVPR 2021 · similarity 0.87)
- Learning the 3D Fauna of the Web (CVPR 2024 · similarity 0.85)
- SAM 3D Animal: Promptable Animal 3D Reconstruction from Images in the Wild (NeurIPS 2026 · 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.
