Research map / Detection and segmentation / 3D detection and driving
RTM3D: Real-time Monocular 3D Detection from Object Keypoints for Autonomous Driving
Peixuan Li, Huaici Zhao, Pengfei Liu, Feidao Cao
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 keypoint FPN (ablation) | RTM3D DLA34 · KITTI val1 AP3D easy | AP3D: 50.14 -> 54.36 | Improved |
| add keypoint FPN (ablation) | RTM3D DLA34 · KITTI val1 AP3D moderate | AP3D: 15.99 -> 16.86 | Improved |
| add keypoint FPN (ablation) | RTM3D DLA34 · KITTI val1 | inference time (s): 0.054 -> 0.055 | No clear effect |
| add orientation to dimension prior (ablation) | RTM3D DLA34 · KITTI val1 AP3D IoU=0.5 easy | AP3D: 51.21 -> 54.18 | Improved |
| add orientation to dimension prior (ablation) | RTM3D DLA34 · KITTI val1 AP3D IoU=0.7 moderate | AP3D: 17.05 -> 16.02 | Hurt |
| add distance prior (ablation) | RTM3D DLA34 · KITTI val1 AP3D IoU=0.7 easy | AP3D: 20.23 -> 20.76 | Improved |
| add keypoint offset (ablation) | RTM3D DLA34 · KITTI val1 AP3D IoU=0.5 hard | AP3D: 35.13 -> 35.84 | Improved |
| add dimension prior to orientation only (ablation) | RTM3D DLA34 · KITTI val1 AP3D IoU=0.7 easy | AP3D: 3.12 -> 20.23 | Improved |
| use all optional components instead of dimension only (ablation) | RTM3D DLA34 · KITTI val1 | inference time (s): 0.058 -> 0.055 | Improved |
| DLA34 backbone instead of ResNet18 | RTM3D · KITTI val1 AP3D IoU=0.7 moderate | AP3D: 14.14 -> 16.86 | Improved |
| DLA34 backbone instead of ResNet18 | RTM3D · KITTI val1 | inference time (s): 0.035 -> 0.055 | Hurt |
| RTM3D vs M3D-RPN on test set | RTM3D DLA34 · KITTI test AP3D easy | AP3D: 14.76 -> 13.61 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- M3D-RPN — M3D-RPN: Monocular 3D Region Proposal Network for Object Detection (ICCV 2019) · better on 3 of 5 shared results
- MonoGRNet — MonoGRNet: A Geometric Reasoning Network for Monocular 3D Object Localization (AAAI 2019) · better on 2 of 2 shared results
- GS3D — GS3D: An Efficient 3D Object Detection Framework for Autonomous Driving (CVPR 2019) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Distance-Normalized Unified Representation for Monocular 3D Object Detection (ECCV 2020 · similarity 0.90)
- Accurate Monocular 3D Object Detection via Color-Embedded 3D Reconstruction for Autonomous Driving (ICCV 2019 · similarity 0.90)
- Monocular 3D Object Detection for Autonomous Driving (CVPR 2016 · similarity 0.90)
- A Depth Complementary Framework for Monocular 3D Object Detection in Autonomous Driving (TPAMI 2026 · similarity 0.90)
- MonoPair: Monocular 3D Object Detection Using Pairwise Spatial Relationships (CVPR 2020 · similarity 0.90)
- MonoDGP: Monocular 3D Object Detection with Decoupled-Query and Geometry-Error Priors (CVPR 2025 · similarity 0.88)
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.
