Research map / Point clouds / Segmentation and detection
Voxel Set Transformer: A Set-to-Set Approach to 3D Object Detection From Point Clouds
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 VoxSeT instead of CT3D | Waymo vehicle detection, LEVEL_1 · Waymo validation, LEVEL_1 vehicle | 3D mAP Overall: 76.02 vs 76.30 | No clear effect |
| Add the CT3D RoI head | VoxSeT · Waymo validation, LEVEL_1 vehicle | 3D mAP Overall: 76.02 -> 77.82 | Helped |
| Add the CT3D RoI head | VoxSeT · Waymo validation, LEVEL_2 vehicle | 3D mAP Overall: 68.16 -> 70.21 | Helped |
| Use VoxSeT instead of CT3D | KITTI validation car detection · KITTI validation, car | 3D mAP Moderate: 86.06 -> 86.71 | Helped |
| Use VoxSeT instead of SECOND | KITTI validation vehicle detection · KITTI validation, vehicle | 3D mAP Easy: 88.61 -> 88.45 | Hurt |
| Use VoxSeT instead of SECOND | KITTI validation pedestrian detection · KITTI validation, pedestrian | 3D mAP Moderate: 52.98 -> 54.74 | Helped |
| Replace ConvFFN with point-wise FFN (ablation) | VoxSeT · KITTI validation, RPN | mAP 11 recall points, Easy: 88.31 -> 70.12 | Hurt |
| Add ConvFFN token interactions (ablation) | VoxSeT · KITTI validation, RPN | mAP 11 recall points, Easy: 70.12 -> 88.31 | Helped |
| Increase latent codes from 4 to 8 (ablation) | VoxSeT VSA modules · KITTI validation, RPN | mAP 11 recall points: 76.63 -> 78.25 | Helped |
| Increase latent codes from 8 to 16 (ablation) | VoxSeT VSA modules · KITTI validation, RPN | mAP 11 recall points: 78.25 -> 78.74 | Helped |
| Replace the PointNet++ backbone with VoxSeT/VSA (ablation) | PointRCNN · KITTI validation, PointRCNN | mAP 11 recall points, Moderate: 78.95 -> 80.14 | Helped |
Closest papers by meaning
- HVPR: Hybrid Voxel-Point Representation for Single-Stage 3D Object Detection (CVPR 2021 · similarity 0.91)
- PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detection (CVPR 2020 · similarity 0.91)
- Voxel R-CNN: Towards High Performance Voxel-based 3D Object Detection (AAAI 2021 · similarity 0.90)
- VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detection (CVPR 2018 · similarity 0.90)
- Voxel Mamba: Group-Free State Space Models for Point Cloud based 3D Object Detection (NeurIPS 2024 · similarity 0.89)
- Voxel or Pillar: Exploring Efficient Point Cloud Representation for 3D Object Detection (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.