Research map / Detection and segmentation / CNN features
Rotation-robust Intersection over Union for 3D Object Detection
Yu Zheng, Danyang Zhang, Sinan Xie, Jiwen Lu, Jie Zhou
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 RIoU loss to l1 regression | F-PointNet v2 · KITTI car 3D detection | AP Moderate: 70.92 -> 72.13 | Improved |
| add RIoU loss to l1 regression | F-PointNet v1 · KITTI car 3D detection | AP Easy: 83.47 -> 84.45 | Improved |
| add RGIoU instead of RIoU | F-PointNet v2 · KITTI car 3D detection | AP Moderate: 72.13 -> 71.71 | Hurt |
| add ArIoU loss to l1 regression | F-PointNet v1 · KITTI car 3D detection | AP Hard: 62.86 -> 60.96 | Hurt |
| add RIoU loss instead of exact IoU loss | PointPillars · KITTI car 3D detection | AP Hard: 75.70 -> 74.04 | Hurt |
| add RIoU loss to l1 | PointPillars · KITTI car 3D detection | AP Easy: 87.29 -> 88.02 | Improved |
| use volume format RIoU(v) instead of BEV RIoU (ablation) | PointPillars · nuScenes car | AP D=0.5: 57.26 -> 61.87 | Improved |
| add RIoU loss to VoteNet | VoteNet · SUN RGB-D | mAP: 57.7 -> 58.8 | Improved |
| replace min with max in intersection estimate (ablation) | RIoU · KITTI car 3D detection | AP Moderate: 71.20 -> 70.79 | Hurt |
| replace min with mean in intersection estimate (ablation) | RIoU · KITTI car 3D detection | AP Moderate: 71.20 -> 69.68 | Hurt |
| add the cosine coefficient to IRIoU (ablation) | RIoU · KITTI car 3D detection | AP Moderate: 69.02 -> 71.20 | Improved |
| set angle coefficient to 2 instead of 5 (ablation) | RIoU · KITTI car 3D detection | AP Moderate: 70.39 -> 71.20 | Improved |
Closest papers by meaning
- Generalized Intersection Over Union: A Metric and a Loss for Bounding Box Regression (CVPR 2019 · similarity 0.88)
- Rethinking IoU-Based Optimization for Single-Stage 3D Object Detection (ECCV 2022 · similarity 0.88)
- Deep Dive Into Gradients: Better Optimization for 3D Object Detection With Gradient-Corrected IoU Supervision (CVPR 2023 · similarity 0.88)
- $\alpha$-IoU: A Family of Power Intersection over Union Losses for Bounding Box Regression (NeurIPS 2021 · similarity 0.86)
- Rethinking Boundary Discontinuity Problem for Oriented Object Detection (CVPR 2024 · similarity 0.86)
- CN-RMA: Combined Network with Ray Marching Aggregation for 3D Indoor Object Detection from Multi-view Images (CVPR 2024 · similarity 0.84)
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.
