Research map / Point clouds / Segmentation and detection
PG-RCNN: Semantic Surface Point Generation for 3D Object Detection
Inyong Koo, Inyoung Lee, Se-Ho Kim, Hee-Seon Kim, Woo-jin Jeon, Changick Kim
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 foreground score supervision L_score to generated points (ablation) | PG-RCNN · KITTI val Car | mAP: 86.10 -> 86.92 | Improved |
| add offset (shape) supervision L_offset (ablation) | PG-RCNN · KITTI val Car | mAP: 86.31 -> 86.92 | Improved |
| add Transformer encoder to RoI grid features (ablation) | PG-RCNN · KITTI val Car | 3D AP R40 Moderate: 82.37 -> 85.22 | Improved |
| predict offsets from grid points instead of RoI center (ablation) | PG-RCNN · KITTI val Car | 3D AP R40 Moderate: 81.94 -> 85.22 | Improved |
| generate semantic points with PG-RCNN instead of SIENet completion (same SECOND RPN) | SIENet re-implementation · KITTI val Car | 3D AP R40 Moderate: 84.45 -> 85.26 | Improved |
| use PG-RCNN instead of heavier point completion detectors | PG-RCNN · KITTI val | Parameters (M): 24.62 -> 2.28 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- PointRCNN — PointRCNN: 3D Object Proposal Generation and Detection From Point Cloud (CVPR 2019) · 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
- LiDAR R-CNN: An Efficient and Universal 3D Object Detector (CVPR 2021 · similarity 0.91)
- Not All Points Are Equal: Learning Highly Efficient Point-Based Detectors for 3D LiDAR Point Clouds (CVPR 2022 · similarity 0.90)
- PointRCNN: 3D Object Proposal Generation and Detection From Point Cloud (CVPR 2019 · similarity 0.90)
- Graph R-CNN: Towards Accurate 3D Object Detection with Semantic-Decorated Local Graph (ECCV 2022 · similarity 0.90)
- PC-RGNN: Point Cloud Completion and Graph Neural Network for 3D Object Detection (AAAI 2021 · similarity 0.90)
- Beyond Virtual Points: Depth-Enhanced LiDAR-only 3D Object Detection with Semi-Supervised Learning (Student Abstract) (AAAI 2025 · similarity 0.89)
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.
