Research map / Point clouds / Shape and surface completion
SketchSampler: Sketch-Based 3D Reconstruction via View-Dependent Depth Sampling
Chenjian Gao, Qian Yu, Lu Sheng, Yi-Zhe Song, Dong Xu
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 sketch decoder (sketch translator) instead of simple encoder only (ablation) | SketchSampler · Synthetic-LineDrawing | reconstruction quality (CD/EMD/FPD/IoU) | Improved |
| add density-guided sampling instead of homogeneous foreground sampling (ablation) | SketchSampler · Synthetic-LineDrawing | reconstruction quality | Improved |
| use ground-truth density maps instead of predicted ones (Ours_real) (ablation) | SketchSampler · Synthetic-LineDrawing | reconstruction quality | No clear effect |
| SketchSampler vs best baseline Sketch2Mesh, CD | SketchSampler · Synthetic-LineDrawing | Chamfer Distance x1e-3: 3.092 -> 0.982 | Improved |
| SketchSampler vs best baseline PCDNet, FPD | SketchSampler · Synthetic-LineDrawing | FPD x10: 1.042 -> 0.580 | Improved |
| SketchSampler vs Sketch2Mesh on Voxel-IoU | SketchSampler · Synthetic-LineDrawing | Voxel-IoU: 0.492 -> 0.578 | Improved |
| SketchSampler vs Sketch2Model on hand-drawn EMD (ShapeNet-Sketch) | SketchSampler · ShapeNet-Sketch | EMD x1e-2: 9.256 -> 9.626 | Hurt |
| SketchSampler vs best baseline on ProSketch CD | SketchSampler · ProSketch-3DChair | Chamfer Distance x1e-3: 5.628 -> 3.868 | Improved |
Closest papers by meaning
- 3D Reconstruction From a Single Sketch via View-Dependent Depth Sampling (TPAMI 2024 · similarity 0.98)
- Sketch2Vox: Learning 3D Reconstruction from a Single Monocular Sketch Image (ECCV 2024 · similarity 0.91)
- Sketch2Mesh: Reconstructing and Editing 3D Shapes From Sketches (ICCV 2021 · similarity 0.90)
- Diff3DS: Generating View-Consistent 3D Sketch via Differentiable Curve Rendering (ICLR 2025 · similarity 0.87)
- SPAR3D: Stable Point-Aware Reconstruction of 3D Objects from Single Images (CVPR 2025 · similarity 0.86)
- Test3R: Learning to Reconstruct 3D at Test Time (NeurIPS 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.
