Research map / Point clouds / Shape and surface completion
Deep Point Cloud Reconstruction
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 joint voxel generation and voxel re-localization reconstruction | Point cloud reconstruction · ScanNet | k-Chamf. (k=1): 2.66 -> 1.91 | Helped |
| Use joint voxel generation and voxel re-localization reconstruction | Point cloud reconstruction · ICL-NUIM | k-Chamf. (k=1): 3.41 -> 1.87 | Helped |
| Use PointRecon with voxel length l_vox=0.0150 | Point cloud reconstruction · ShapeNet-Part | k-Chamf. (k=1): 1.14 -> 0.90 | Helped |
| Add amplified positional encoding instead of using no positional encoding (ablation) | PointRecon · ShapeNet-Part | Chamf.: 1.231 -> 1.195 | Helped |
| Replace γ_PE with amplified positional encoding (ablation) | PointRecon · ShapeNet-Part | Chamf.: 1.224 -> 1.195 | Helped |
| Replace PE* with amplified positional encoding (ablation) | PointRecon · ShapeNet-Part | Chamf.: 1.226 -> 1.195 | Helped |
| Add self-attention and cross-attention to the first-stage voxel network (ablation) | PointRecon · ShapeNet-Part | Chamf.: 1.42 -> 1.195 | Helped |
| Add cross-attention to the self-attention variant (ablation) | PointRecon · ShapeNet-Part | Chamf.: 1.209 -> 1.195 | Helped |
| Add self-attention to the cross-attention variant (ablation) | PointRecon · ShapeNet-Part | Chamf.: 1.223 -> 1.195 | Helped |
Closest papers by meaning
- Arbitrary-Scale Point Cloud Upsampling by Voxel-Based Network with Latent Geometric-Consistent Learning (AAAI 2024 · similarity 0.88)
- EdgeMovingNet: Edge-preserving Point Cloud Reconstruction via Joint Geometry Features (CVPR 2025 · similarity 0.88)
- Depth-Guided Robust and Fast Point Cloud Fusion NeRF for Sparse Input Views (AAAI 2024 · similarity 0.87)
- LinNet: Linear Network for Efficient Point Cloud Representation Learning (NeurIPS 2024 · similarity 0.87)
- Learning Continuous Implicit Field with Local Distance Indicator for Arbitrary-Scale Point Cloud Upsampling (AAAI 2024 · similarity 0.87)
- DC-PCN: Point Cloud Completion Network with Dual-Codebook Guided Quantization (AAAI 2025 · 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.