Research map / Detection and segmentation / 3D detection and driving
DOPS: Learning to Detect 3D Objects and Predict Their 3D Shapes
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 DOPS instead of DeepVote | 3D object detection · ScanNetV2 validation | mAP@0.25: 58.6 -> 63.7 | Helped |
| Use DOPS instead of DeepVote | 3D object detection · ScanNetV2 validation | mAP@0.5: 33.5 -> 38.2 | Helped |
| Use DOPS instead of StarNet | 3D object detection · Waymo Open Dataset | mAP@0.7: 53.0 -> 56.4 | Helped |
| Add graph convolution over neighborhood predictions (ablation) | DOPS · Waymo Open Dataset | mAP@0.7: 54.5 -> 56.4 | Helped |
| Use dynamic classification loss instead of regular classification loss (ablation) | DOPS · Waymo Open Dataset | mAP@0.7: 53.1 -> 56.4 | Helped |
| Use farthest and highest object sampling instead of NMS (ablation) | DOPS · Waymo Open Dataset | mAP@0.7: 55.7 -> 56.4 | Helped |
| Add shape prediction (ablation) | DOPS detection pipeline · Waymo Open Dataset | detection precision | No clear effect |
| Remove internal points when training the ShapeNet prior (ablation) | DOPS shape prior · Shape prediction on Waymo | shape prediction quality | Helped |
| Add inside/outside samples along rays from observed points toward the object center (ablation) | DOPS shape fitting · Waymo Open Dataset | recovered shape quality | Helped |
| Decrease the ray-sample distance δ from 0.5 to 0.1 (ablation) | DOPS shape fitting · Waymo Open Dataset | recovered shape quality: 0.5 -> 0.1 | Helped |
Closest papers by meaning
- LaserNet: An Efficient Probabilistic 3D Object Detector for Autonomous Driving (CVPR 2019 · similarity 0.91)
- Fully Sparse 3D Object Detection (NeurIPS 2022 · similarity 0.90)
- VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detection (CVPR 2018 · similarity 0.90)
- Pseudo-LiDAR From Visual Depth Estimation: Bridging the Gap in 3D Object Detection for Autonomous Driving (CVPR 2019 · similarity 0.90)
- NBA3D: Neighbor-Based Confidence Adjustment for 3D Rare Object Detection Using LiDAR (AAAI 2025 · similarity 0.89)
- SAFDNet: A Simple and Effective Network for Fully Sparse 3D Object Detection (CVPR 2024 · 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.