Research map / 3D vision / Face and shape
Learning Manifold Patch-Based Representations of Man-Made Shapes
Read the paperCheck 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 |
|---|---|---|---|
| Use Coons patches instead of a triangle mesh | Sketch-based reconstruction · airplane test views | degrees of freedom: 7398 -> 816 | Helped |
| Use Coons patches instead of a triangle mesh | Sketch-based reconstruction · car test views | degrees of freedom: 7398 -> 816 | Helped |
| Use normal alignment loss (ablation) | Patch-based reconstruction · airplanes test set | normal loss: 1.09 -> 0.725 | Helped |
| Use normal alignment loss (ablation) | Patch-based reconstruction · knives test set | normal loss: 1.12 -> 0.635 | Helped |
| Use collision detection loss (ablation) | Patch-based reconstruction · airplanes test set | intersecting patch pairs: 3.09 -> 0.0220 | Helped |
| Use collision detection loss (ablation) | Patch-based reconstruction · knives test set | intersecting patch pairs: 0.0263 -> 0 | Helped |
| Use patch flatness loss (ablation) | Patch-based reconstruction · airplanes test set | flatness loss: 0.000254 -> 0.0000732 | Helped |
| Use patch flatness loss (ablation) | Patch-based reconstruction · knives test set | flatness loss: 0.00132 -> 0.0000950 | Helped |
| Use template normal loss (ablation) | Patch-based reconstruction · airplanes test set | template normal loss: 0.889 -> 0.792 | Helped |
| Use template normal loss (ablation) | Patch-based reconstruction · knives test set | template normal loss: 1.03 -> 0.669 | Helped |
| Use category-specific templates instead of generic sphere templates (ablation) | Patch-based reconstruction · synthetic sketch reconstruction | reconstruction quality | Helped |
| Use more patches in the sphere template (ablation) | Patch-based reconstruction · airplane model | captured detail | Helped |
Closest papers by meaning
- DeepCurrents: Learning Implicit Representations of Shapes With Boundaries (CVPR 2022 · similarity 0.88)
- SurfNet: Generating 3D Shape Surfaces Using Deep Residual Networks (CVPR 2017 · similarity 0.88)
- Manifold Learning in Quotient Spaces (CVPR 2018 · similarity 0.88)
- Geometry-Aware Network for Non-Rigid Shape Prediction From a Single View (CVPR 2018 · similarity 0.88)
- DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation (CVPR 2019 · similarity 0.88)
- Enhancing Implicit Shape Generators Using Topological Regularizations (ICML 2024 · similarity 0.87)
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.