Research map / Detection and segmentation / 3D detection and driving
Stereo R-CNN Based 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 |
|---|---|---|---|
| add channel concatenation instead of element-wise mean for left-right feature fusion (ablation) | Stereo R-CNN · KITTI validation set | Stereo AR (300 proposals): 74.60 -> 75.51 | Helped |
| add stereo flip augmentation (ablation) | Stereo R-CNN · KITTI validation set, moderate | AP_3d (IoU=0.7): 21.52 -> 28.74 | Helped |
| add uncertainty weighting to the multiple losses (ablation) | Stereo R-CNN · KITTI validation set, moderate | AP_3d (IoU=0.7): 21.52 -> 30.36 | Helped |
| add perspective keypoint constraint to the 3D box estimator (ablation) | Stereo R-CNN · KITTI validation set, moderate | AP_bv (IoU=0.7): 40.44 -> 48.30 | Helped |
| add perspective keypoint constraint to the 3D box estimator (ablation) | Stereo R-CNN · KITTI validation set, moderate | AP_3d (IoU=0.7): 30.29 -> 36.69 | Helped |
| add dense photometric alignment (ablation) | Stereo R-CNN · KITTI validation set, moderate | AP_3d (IoU=0.7): 7.75 -> 32.00 | Helped |
| add 3D box rectification after dense alignment (ablation) | Stereo R-CNN with dense alignment · KITTI validation set, moderate | AP_3d (IoU=0.7): 32.00 -> 36.69 | Helped |
| use channel concatenation rather than element-wise mean for left-right feature fusion (ablation) | Stereo R-CNN · KITTI validation set, moderate | Stereo AP_2d (IoU=0.7): 88.24 -> 88.27 | No clear effect |
Closest papers by meaning
- IDA-3D: Instance-Depth-Aware 3D Object Detection From Stereo Vision for Autonomous Driving (CVPR 2020 · similarity 0.90)
- Disp R-CNN: Stereo 3D Object Detection via Shape Prior Guided Instance Disparity Estimation (CVPR 2020 · similarity 0.90)
- LiDAR R-CNN: An Efficient and Universal 3D Object Detector (CVPR 2021 · similarity 0.89)
- GS3D: An Efficient 3D Object Detection Framework for Autonomous Driving (CVPR 2019 · similarity 0.89)
- Monocular 3D Object Detection for Autonomous Driving (CVPR 2016 · similarity 0.89)
- DriveGEN: Generalized and Robust 3D Detection in Driving via Controllable Text-to-Image Diffusion Generation (CVPR 2025 · similarity 0.87)
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.