Research map / Point clouds / Shape and surface completion
SAL: Sign Agnostic Learning of Shapes From Raw Data
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 test-time latent-vector optimization for 800 iterations (ablation) | SAL · D-Faust test poses | Chamfer distance to registrations, median (* 10^3): 0.37 -> 0.16 | Helped |
| add test-time latent-vector optimization for 800 iterations (ablation) | SAL · D-Faust test poses | Chamfer distance to scans, median (* 10^3): 0.18 -> 0.09 | Helped |
| add test-time latent-vector optimization for 800 iterations (ablation) | SAL · D-Faust test humans | Chamfer distance to registrations, median (* 10^3): 0.75 -> 0.3 | Helped |
| add test-time latent-vector optimization for 800 iterations (ablation) | SAL · D-Faust test humans | Chamfer distance to scans, median (* 10^3): 0.34 -> 0.14 | Helped |
| learn a shared shape space with the SAL VAE rather than reconstruct each scan independently (ablation) | SAL · D-Faust training scans | Chamfer distance to registrations, median (* 10^3): 0.17 -> 0.09 | Helped |
| learn a shared shape space with the SAL VAE rather than reconstruct each scan independently (ablation) | SAL · D-Faust training scans | Chamfer distance to scans, median (* 10^3): 0.08 -> 0.06 | No clear effect |
Closest papers by meaning
- SALD: Sign Agnostic Learning with Derivatives (ICLR 2021 · similarity 0.94)
- Sign-Agnostic Implicit Learning of Surface Self-Similarities for Shape Modeling and Reconstruction From Raw Point Clouds (CVPR 2021 · similarity 0.89)
- Shape as Line Segments: Accurate and Flexible Implicit Surface Representation (ICLR 2025 · similarity 0.87)
- LISR: Learning Linear 3D Implicit Surface Representation Using Compactly Supported Radial Basis Functions (AAAI 2024 · similarity 0.86)
- Small Steps and Level Sets: Fitting Neural Surface Models with Point Guidance (CVPR 2024 · similarity 0.86)
- A Lightweight UDF Learning Framework for 3D Reconstruction Based on Local Shape Functions (CVPR 2025 · 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.