Research map / Detection and segmentation / 3D detection and driving
Progressive Coordinate Transforms for Monocular 3D Object Detection
Li Wang, Li Zhang, Yi Zhu, Zhi Zhang, Tong He, Mu Li, Xiangyang Xue
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 localization boosting (T=3) to PatchNet (ablation) | PatchNet · KITTI val AP3D Mod. | AP3D: 25.88 -> 26.78 | Improved |
| add confidence-aware loss to localization boosting (ablation) | PatchNet + boosting · KITTI val AP3D Mod. | AP3D: 26.78 -> 27.24 | Improved |
| add global context encoding to boosted model (ablation) | PatchNet + boosting · KITTI val AP3D Mod. | AP3D: 26.78 -> 27.12 | Improved |
| combine CLB and GCE (full PCT) (ablation) | PatchNet · KITTI val AP3D Mod. | AP3D: 25.88 -> 27.53 | Improved |
| increase boosting iterations from 3 to 4 or 5 (ablation) | PatchNet boosting · KITTI val AP3D Mod. | AP3D: 26.78 -> 26.77 (T=4), 26.64 (T=5) | Hurt |
| fuse GCE into all networks instead of only G (ablation) | PatchNet + boosting · KITTI val AP3D Mod. | AP3D: 27.12 -> 27.07 | No clear effect |
| fuse GCE into F rather than G (ablation) | PatchNet + boosting · KITTI val AP3D Mod. | AP3D: 27.12 -> 27.08 | Hurt |
| add CLB and GCE to Pseudo-LiDAR | Pseudo-LiDAR · KITTI val AP3D Mod. | AP3D: 23.04 -> 24.43 | Improved |
| add PCT to PatchNet on Waymo | PatchNet · Waymo val LEVEL_1 IoU=0.7 | 3D mAP / mAPH: 0.39 / 0.37 -> 0.89 / 0.88 | Improved |
| add PCT to PatchNet with PSMNet depth | PatchNet · KITTI val AP3D Mod. | AP3D: 47.30 -> 48.27 | Improved |
| add PCT to Pseudo-LiDAR with PSMNet depth | Pseudo-LiDAR · KITTI val AP3D Mod. | AP3D: 42.01 -> 45.31 | Improved |
| add PCT to PatchNet on Pedestrian | PatchNet · KITTI val Pedestrian AP3D Mod. | AP3D: 12.17 -> 15.31 | Improved |
Closest papers by meaning
- MonoDistill: Learning Spatial Features for Monocular 3D Object Detection (ICLR 2022 · similarity 0.88)
- MonoRUn: Monocular 3D Object Detection by Reconstruction and Uncertainty Propagation (CVPR 2021 · similarity 0.88)
- Monocular 3D Object Detection Leveraging Accurate Proposals and Shape Reconstruction (CVPR 2019 · similarity 0.88)
- PTT: Point-Trajectory Transformer for Efficient Temporal 3D Object Detection (CVPR 2024 · similarity 0.87)
- MonoPlace3D: Learning 3D-Aware Object Placement for 3D Monocular Detection (CVPR 2025 · similarity 0.86)
- Leveraging Imagery Data with Spatial Point Prior for Weakly Semi-supervised 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.
