Research map / 3D vision / Face and shape
Geometric Granularity Aware Pixel-To-Mesh
Yue Shi, Bingbing Ni, Jinxian Liu, Dingyi Rong, Ye Qian, Wenjun Zhang
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 |
|---|---|---|---|
| add Geometry Structure Extractor (ablation) | full model · ShapeNet 5 categories mean | CD: 0.5278 -> 0.4894 | Improved |
| add Geometry Structure Extractor (ablation) | full model · ShapeNet 5 categories mean | F1: 68.014 -> 71.202 | Improved |
| add fine-grained edit module (ablation) | full model · ShapeNet 5 categories mean | CD: 0.5928 -> 0.4894 | Improved |
| add fine-grained edit module (ablation) | full model · ShapeNet 5 categories mean | F1: 66.168 -> 71.202 | Improved |
| add Geometry Structure Extractor (ablation) | full model · ShapeNet monitor | F1: 59.89 -> 71.20 | Improved |
| add edit module (ablation) | full model · ShapeNet monitor | CD: 0.539 -> 0.517 | Mixed |
| add GSE to Pixel2Mesh (ablation) | Pixel2Mesh · ShapeNet average | CD: 0.756 -> 0.659 | Improved |
| add edit module to Pixel2Mesh (ablation) | Pixel2Mesh · ShapeNet average | CD: 0.756 -> 0.603 | Improved |
| add GSE to Pixel2Mesh (ablation) | Pixel2Mesh · ShapeNet average | F1: 55.56 -> 57.94 | Improved |
| add edit module to Pixel2Mesh (ablation) | Pixel2Mesh · ShapeNet average | F1: 55.56 -> 59.87 | Improved |
| replace DeepMesh topology modification with full framework | DeepMesh · ShapeNet 5 categories mean | CD: 0.6212 -> 0.4894 | Improved |
| increase number of key points (more points) (ablation) | GSE without edit module · ShapeNet | CD | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Pixel2Mesh — Pixel2Mesh: Generating 3D Mesh Models from Single RGB Images (ECCV 2018) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- QUADify: Extracting Meshes with Pixel-level Details and Materials from Images (CVPR 2024 · similarity 0.87)
- 3D-PSSIM: Projective Structural Similarity for 3D Mesh Quality Assessment Robust to Topological Irregularities (TPAMI 2024 · similarity 0.86)
- Material Magic Wand: Material-Aware Grouping of 3D Parts in Untextured Meshes (CVPR 2026 · similarity 0.86)
- VGGTFace: Topologically Consistent Facial Geometry Reconstruction in the Wild (AAAI 2026 · similarity 0.86)
- SILSA: Sliding-Window Slice Latents for Topology-Preserving High-Resolution 3D Generation (NeurIPS 2026 · similarity 0.85)
- Mesh-RFT: Enhancing Mesh Generation via Fine-grained Reinforcement Fine-Tuning (NeurIPS 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.
