Research map / Point clouds / Shape and surface completion
TopNet: Structural Point Cloud Decoder
Check 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 the TopNet tree decoder instead of the strongest baseline decoder | ShapeNet completion with the 2-stage PointNet encoder · ShapeNet completion, 2048 points | Chamfer distance (×10^4): 14.72 -> 9.72 | Helped |
| Use the 2-stage encoder instead of the 1-stage PointNet encoder (ablation) | TopNet · ShapeNet completion, 2048 points | Chamfer distance (×10^4): 12.97 -> 9.72 | Helped |
| Increase the output point count from 2048 to 4096 (ablation) | TopNet · ShapeNet completion | Chamfer distance (×10^4): 9.72 -> 8.83 | Helped |
| Increase the output point count from 4096 to 8192 (ablation) | TopNet · ShapeNet completion | Chamfer distance (×10^4): 8.83 -> 7.20 | Helped |
| Increase the output point count from 8192 to 16384 (ablation) | TopNet · ShapeNet completion | Chamfer distance (×10^4): 7.20 -> 6.50 | Helped |
| Vary the number of tree levels (ablation) | TopNet · ShapeNet completion, 2048 output points | Chamfer distance | Mixed |
| Vary the node feature dimension (ablation) | TopNet · ShapeNet completion, 2048 output points | Chamfer distance | No clear effect |
| Use more network parameters (ablation) | TopNet · ShapeNet completion | Chamfer distance | No clear effect |
Closest papers by meaning
- Point Cloud Completion by Skip-Attention Network With Hierarchical Folding (CVPR 2020 · similarity 0.90)
- EditVAE: Unsupervised Parts-Aware Controllable 3D Point Cloud Shape Generation (AAAI 2022 · similarity 0.90)
- DC-PCN: Point Cloud Completion Network with Dual-Codebook Guided Quantization (AAAI 2025 · similarity 0.89)
- PointCFormer: A Relation-Based Progressive Feature Extraction Network for Point Cloud Completion (AAAI 2025 · similarity 0.89)
- Orthogonal Dictionary Guided Shape Completion Network for Point Cloud (AAAI 2024 · similarity 0.89)
- PointAttN: You Only Need Attention for Point Cloud Completion (AAAI 2024 · similarity 0.88)
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.