Research map / Point clouds / Point cloud classification
VV-Net: Voxel VAE Net With Group Convolutions for Point Cloud Segmentation
Hsien-Yu Meng, Lin Gao, Yu-Kun Lai, Dinesh Manocha
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 group convolution on top of RBF-VAE (ablation) | VV-Net · ShapeNet part segmentation | instance mIoU: 86.1 -> 87.4 | Improved |
| add RBF-VAE on top of group convolution (instead of {0,1} voxels) (ablation) | VV-Net · ShapeNet part segmentation | instance mIoU: 86.0 -> 87.4 | Improved |
| use plain {0,1} occupancy VAE instead of RBF-VAE (ablation) | VAE voxel encoder · ShapeNet part segmentation | training convergence | Hurt |
| add group convolution on top of RBF-VAE (ablation) | VV-Net · S3DIS semantic segmentation | mean IoU: 75.40 -> 78.22 | Improved |
| add RBF-VAE on top of group convolution (instead of {0,1} voxels) (ablation) | VV-Net · S3DIS semantic segmentation | mean IoU: 68.70 -> 78.22 | Improved |
| add group convolution on top of RBF-VAE (ablation) | VV-Net · S3DIS semantic segmentation (AP at IoU 0.5) | mean AP@0.5: 79.00 -> 82.17 | Improved |
| add RBF-VAE on top of group convolution (instead of {0,1} voxels) (ablation) | VV-Net · S3DIS semantic segmentation (AP at IoU 0.5) | mean AP@0.5: 72.66 -> 82.17 | Improved |
| use G-CNN instead of traditional CNN (ablation) | VV-Net · S3DIS semantic segmentation | mean IoU: 67.61 -> 75.40 | Improved |
| use RBF-VAE latent voxels instead of RBF grids (ablation) | VV-Net · S3DIS semantic segmentation | mean IoU: 64.13 -> 75.40 | Improved |
| use RBF-VAE latent voxels instead of finer 64x64x128 grid (ablation) | VV-Net · S3DIS semantic segmentation | mean IoU: 70.00 -> 75.40 | Improved |
| use VAE instead of plain AE (ablation) | VV-Net · S3DIS semantic segmentation | mean IoU: 69.60 -> 75.40 | Improved |
| use Gaussian RBF kernel instead of inverse quadratic kernel (ablation) | VV-Net · S3DIS semantic segmentation | mean IoU: 65.04 -> 78.22 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- SGPN — SGPN: Similarity Group Proposal Network for 3D Point Cloud Instance Segmentation (CVPR 2018) · better on 1 of 1 shared result
- SO-Net — SO-Net: Self-Organizing Network for Point Cloud Analysis (CVPR 2018) · better on 1 of 1 shared result
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
- Deep FusionNet for Point Cloud Semantic Segmentation (ECCV 2020 · similarity 0.89)
- Fully-Convolutional Point Networks for Large-Scale Point Clouds (ECCV 2018 · similarity 0.88)
- PointCNN++: Performant Convolution on Native Points (CVPR 2026 · similarity 0.88)
- RPVNet: A Deep and Efficient Range-Point-Voxel Fusion Network for LiDAR Point Cloud Segmentation (ICCV 2021 · similarity 0.88)
- Arbitrary-Scale Point Cloud Upsampling by Voxel-Based Network with Latent Geometric-Consistent Learning (AAAI 2024 · similarity 0.87)
- Voxel or Pillar: Exploring Efficient Point Cloud Representation for 3D Object Detection (AAAI 2024 · 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.
