Research map / Point clouds / Segmentation and detection
Weakly Supervised 3D Object Detection from Lidar Point Cloud
Qinghao Meng, Wenguan Wang, Tianfei Zhou, Jianbing Shen, Luc Van Gool, Dengxin Dai
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 |
|---|---|---|---|
| use weak click annotation (500 scenes + 534 precise instances) instead of 500 fully labeled scenes | PointRCNN · KITTI val Car 3D Box@0.7 Moderate | AP: 65.50 -> 75.10 | Improved |
| use weak annotation instead of 500 fully labeled scenes | PointPillars · KITTI val Car 3D Box@0.7 Moderate | AP: 60.75 -> 75.10 | Improved |
| weak supervision (about 3% of precise instances) versus full supervision with 3,712 scenes | PointRCNN · KITTI val Car 3D Box@0.7 Moderate | AP: 77.67 -> 75.10 | Hurt |
| weak supervision vs full supervision on KITTI test | STD · KITTI test Car 3D Box@0.7 Moderate | AP: 79.71 -> 69.64 | Hurt |
| train on pedestrians with 951 click-labeled scenes and 515 instances | weakly supervised detector · KITTI val Pedestrian 3D Box@0.5 Moderate | AP: 61.06 -> 69.96 | Improved |
| add precise BEV annotations instead of weak click BEV (ablation) | weakly supervised detector · KITTI val Car 3D Box@0.7 Moderate | AP: 75.10 -> 75.13 | No clear effect |
| add more weakly labeled scenes (3,712) with 534 precise instances (ablation) | weakly supervised detector · KITTI val Car 3D Box@0.7 Moderate | AP: 75.10 -> 76.04 | Improved |
| add 1,852 weakly labeled scenes with 25% precise instances (ablation) | weakly supervised detector · KITTI val Car 3D Box@0.7 Moderate | AP: 75.10 -> 76.78 | Improved |
| add 3,712 weakly labeled scenes with 25% precise instances (ablation) | weakly supervised detector · KITTI val Car 3D Box@0.7 Moderate | AP: 75.10 -> 77.62 | Improved |
| train PointRCNN on our automatic labels instead of manual labels | PointRCNN · KITTI val Car 3D Box@0.7 Moderate | AP: 77.67 -> 74.54 | Hurt |
| train PointRCNN on our active labels instead of manual labels | PointRCNN · KITTI val Car 3D Box@0.7 Moderate | AP: 77.67 -> 76.08 | Hurt |
| train PointPillars on our active labels instead of manual labels | PointPillars · KITTI val Car 3D Box@0.7 Moderate | AP: 76.07 -> 75.03 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 3 on most shared results.
- PointPillars — PointPillars: Fast Encoders for Object Detection From Point Clouds (CVPR 2019) · better on 1 of 2 shared results
- PointRCNN — PointRCNN: 3D Object Proposal Generation and Detection From Point Cloud (CVPR 2019) · better on 1 of 2 shared results
- VoxelNet — VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detection (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
- Towards A Weakly Supervised Framework for 3D Point Cloud Object Detection and Annotation (TPAMI 2021 · similarity 0.98)
- Weakly Supervised 3D Object Detection via Multi-Level Visual Guidance (ECCV 2024 · similarity 0.93)
- WeakPCSOD: Overcoming the Bias of Box Annotations for Weakly Supervised Point Cloud Salient Object Detection (AAAI 2024 · similarity 0.90)
- Leveraging Imagery Data with Spatial Point Prior for Weakly Semi-supervised 3D Object Detection (AAAI 2024 · similarity 0.90)
- VSRD: Instance-Aware Volumetric Silhouette Rendering for Weakly Supervised 3D Object Detection (CVPR 2024 · similarity 0.90)
- Weak-to-Strong 3D Object Detection with X-Ray Distillation (CVPR 2024 · similarity 0.90)
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.
