Research map / Point clouds / Point cloud classification
SPLATNet: Sparse Lattice Networks for Point Cloud Processing
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 |
|---|---|---|---|
| add joint 2D image features to 3D point-cloud processing | SPLATNet3D · RueMonge2014 point cloud labeling | Average IoU: 65.4 -> 69.8 | Helped |
| add 3D information to image segmentation | DeepLab2D · RueMonge2014 multi-view image labeling | Average IoU: 1.3 IoU | Helped |
| add rendered 2D information to 3D point-cloud part segmentation | SPLATNet3D · ShapeNet Part | Class average mIoU: 1.7 | Helped |
| add rendered 2D information to 3D point-cloud part segmentation | SPLATNet3D · ShapeNet Part | Instance average mIoU: 0.8 | Helped |
| add a BCL using XYZ and normal lattice features (ablation) | SPLATNet3D · ShapeNet Part | Class and instance average mIoU: 0.2 IoU | Helped |
| omit the initial 1×1 CONV layer (ablation) | SPLATNet3D · RueMonge2014 facade segmentation | Average IoU | No clear effect |
Closest papers by meaning
- SplAttN: Bridging 2D and 3D with Gaussian Soft Splatting and Attention for Point Cloud Completion (ICML 2026 · similarity 0.88)
- PointCNN++: Performant Convolution on Native Points (CVPR 2026 · similarity 0.88)
- Mining Point Cloud Local Structures by Kernel Correlation and Graph Pooling (CVPR 2018 · similarity 0.88)
- LinNet: Linear Network for Efficient Point Cloud Representation Learning (NeurIPS 2024 · similarity 0.87)
- UniPre3D: Unified Pre-training of 3D Point Cloud Models with Cross-Modal Gaussian Splatting (CVPR 2025 · similarity 0.86)
- Sonata: Self-Supervised Learning of Reliable Point Representations (CVPR 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.