Research map / Detection and segmentation / 3D detection and driving
Pseudo-LiDAR From Visual Depth Estimation: Bridging the Gap in 3D Object Detection for Autonomous Driving
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 |
|---|---|---|---|
| Represent stereo depth as pseudo-LiDAR rather than frontal-view depth features (ablation) | AVOD with PSMNet* · KITTI val, car, moderate, IoU 0.7 | AP_3D: 6.6 -> 45.3 | Helped |
| Represent stereo depth as pseudo-LiDAR rather than frontal-view depth features | AVOD with DispNet · KITTI val, car, moderate, IoU 0.7 | AP_3D: 9.8 -> 27.0 | Helped |
| Use PSMNet* trained on the detection training set instead of PSMNet trained on KITTI stereo images | AVOD with pseudo-LiDAR · KITTI val, car, moderate, IoU 0.7 | AP_3D: 27.4 -> 45.3 | Helped |
| Use PSMNet* rather than PSMNet for disparity estimation | F-PointNet with pseudo-LiDAR · KITTI val, car, moderate, IoU 0.7 | AP_3D: 26.7 -> 39.8 | Helped |
| Use DispNet-C rather than PSMNet for disparity estimation | F-PointNet with pseudo-LiDAR · KITTI val, car, moderate, IoU 0.7 | AP_3D: 26.7 -> 29.2 | Helped |
| Replace DispNet-S with DispNet-C for disparity estimation | AVOD with pseudo-LiDAR · KITTI val, car, moderate, IoU 0.7 | AP_3D: 27.0 -> 26.2 | Hurt |
| Use different stereo disparity estimators | AVOD with pseudo-LiDAR · KITTI val, car, moderate, IoU 0.7 | AP_3D: 27.0; 26.2; 27.4; 28.3 | Mixed |
| Over-smooth the depth estimates with an 11×11 average kernel (ablation) | AVOD with pseudo-LiDAR · KITTI val, car, moderate, IoU 0.7 | AP_3D: 45.3 -> 37.8 | Hurt |
| Over-smooth the depth estimates with an 11×11 average kernel (ablation) | F-PointNet with pseudo-LiDAR · KITTI val, car, moderate, IoU 0.7 | AP_3D: 39.8 -> 31.6 | Hurt |
Closest papers by meaning
- Pseudo-LiDAR++: Accurate Depth for 3D Object Detection in Autonomous Driving (ICLR 2020 · similarity 0.96)
- End-to-End Pseudo-LiDAR for Image-Based 3D Object Detection (CVPR 2020 · similarity 0.94)
- Pseudo-Stereo for Monocular 3D Object Detection in Autonomous Driving (CVPR 2022 · similarity 0.93)
- LaserNet: An Efficient Probabilistic 3D Object Detector for Autonomous Driving (CVPR 2019 · similarity 0.92)
- LiDAR R-CNN: An Efficient and Universal 3D Object Detector (CVPR 2021 · similarity 0.91)
- 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.