Research map / 3D vision / Face and shape
Deep Marching Tetrahedra: a Hybrid Representation for High-Resolution 3D Shape Synthesis
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 DMTet instead of ConvOnet | coarse voxel to high-res shape · animal super resolution | L1 Chamfer: 2.41 -> 2.19 | Helped |
| add adversarial loss (ablation) | DMTet · animal super resolution | L1 Chamfer: 2.20 -> 2.19 | Helped |
| add adversarial loss (ablation) | DMTet · user study | preference better looking: 71% wins | Helped |
| use DMTet instead of ConvOnet | point cloud reconstruction · ShapeNet | mean Chamfer L1: 0.95 -> 0.77 | Helped |
| use DMTet instead of ConvOnet | point cloud reconstruction · ShapeNet | inference time ms: 866 -> 129 | Helped |
| add learned vertex deformation to SDF prediction (ablation) | DMTet fixed grid · ShapeNet | mean Chamfer L1: 0.91 -> 0.81 | Helped |
| add volume subdivision (ablation) | DMTet · ShapeNet | mean Chamfer L1: 0.79 -> 0.77 | Helped |
| add surface subdivision (ablation) | DMTet · ShapeNet | mean Chamfer L1: 0.78 -> 0.77 | Helped |
| add volume subdivision (ablation) | DMTet_B · user study | preference better looking: 78% wins | Helped |
| add surface subdivision (ablation) | DMTet_V · user study | preference better looking: 62% wins | Helped |
| add volume subdivision (ablation) | DMTet · ShapeNet | inference time ms: 52 -> 67 | Hurt |
Closest papers by meaning
- Learning Dynamic Tetrahedra for High-Quality Talking Head Synthesis (CVPR 2024 · similarity 0.89)
- Tetrahedron Splatting for 3D Generation (NeurIPS 2024 · similarity 0.88)
- CraftsMan3D: High-fidelity Mesh Generation with 3D Native Diffusion and Interactive Geometry Refiner (CVPR 2025 · similarity 0.87)
- GALA: Geometry-Aware Local Adaptive Grids for Detailed 3D Generation (ICLR 2025 · similarity 0.86)
- ShapeShifter: 3D Variations Using Multiscale and Sparse Point-Voxel Diffusion (CVPR 2025 · similarity 0.86)
- MeshGPT: Generating Triangle Meshes with Decoder-Only Transformers (CVPR 2024 · 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.