Research map / 3D vision / Depth and stereo
A Benchmark and Evaluation of Non-Rigid Structure from Motion
Sebastian Hoppe Nesgaard Jensen, Mads Emil Brix Doest, Henrik Aanæs, Alessio Del Bue
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 |
|---|---|---|---|
| switch from orthographic to full perspective camera model | 18 NRSfM methods · NRSfM benchmark | mean reconstruction error (mm): 50.45 -> 57.66 | Hurt |
| add occlusion-based missing data | MetricProj · NRSfM benchmark | mean reconstruction error (mm): increase of 5.85mm | Hurt |
| add occlusion-based missing data | KSTA · NRSfM benchmark | mean reconstruction error (mm): 9.65mm | Hurt |
| add occlusion-based missing data | CSF2 · NRSfM benchmark | mean reconstruction error (mm): 18.15mm | Hurt |
| add occlusion-based missing data | CSF · NRSfM benchmark | mean reconstruction error (mm): 13.49mm | Hurt |
| add occlusion-based missing data | ScalableSurface (DCT basis) · NRSfM benchmark | mean reconstruction error (mm): 16.52mm | Hurt |
| add occlusion-based missing data | BALM · NRSfM benchmark | mean reconstruction error (mm): decreases by 3.33mm | Improved |
| add occlusion-based missing data to in-plane deformations (Tearing, Stretching, Articulated) | NRSfM methods · NRSfM benchmark | mean reconstruction error (mm): 3.96mm, 4.65mm and 8.38mm | Hurt |
| add occlusion-based missing data to volumetric deformations (Deflation, Bending) | NRSfM methods · NRSfM benchmark | mean reconstruction error (mm): 12.27mm and 13.47mm | Hurt |
| use Circle camera path instead of Zigzag | NRSfM methods · NRSfM benchmark | mean reconstruction error (mm): 47.29 -> 61.18 | Hurt |
| use Tricky versus Circle camera path (designed to be harder) | NRSfM methods · NRSfM benchmark | mean reconstruction error (mm): differ by 3.08mm | No clear effect |
| physical deformation priors (MDH, SoftInext) instead of statistical models | MDH / SoftInext · NRSfM benchmark | mean reconstruction error (mm): 70.34 and 61.94 vs MultiBody 29.36 | Hurt |
Closest papers by meaning
- Structure From Recurrent Motion: From Rigidity to Recurrency (CVPR 2018 · similarity 0.89)
- Scalable Dense Non-Rigid Structure-From-Motion: A Grassmannian Perspective (CVPR 2018 · similarity 0.89)
- Reconstruct as Far as You Can: Consensus of Non-Rigid Reconstruction from Feasible Regions (TPAMI 2019 · similarity 0.89)
- Deep Non-Rigid Structure-from-Motion Revisited: Canonicalization and Sequence Modeling (AAAI 2025 · similarity 0.87)
- Non-Rigid Structure-from-Motion: Temporally-Smooth Procrustean Alignment and Spatially-Variant Deformation Modeling (CVPR 2024 · similarity 0.87)
- Structure-from-Motion with a Non-Parametric Camera Model (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.
