Research map / Detection and segmentation / 3D detection and driving
Distance-Normalized Unified Representation for Monocular 3D Object Detection
Xuepeng Shi, Zhixiang Chen, Tae-Kyun Kim
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 low-resolution depth image to normalize distance targets (ablation) | UR3D baseline · KITTI val1 car, AP3D Easy | AP|R40: 11.26 -> 18.57 | Improved |
| add low-resolution depth image to normalize distance targets (ablation) | UR3D baseline · KITTI val1 car, APBEV Easy | AP|R40: 18.26 -> 27.65 | Improved |
| add distance-guided NMS (K = 1) (ablation) | UR3D + LR depth · KITTI val1 car, AP3D Easy | AP|R40: 18.57 -> 19.75 | Improved |
| add distance-guided NMS (K = 1) (ablation) | UR3D + LR depth · KITTI val1 car, APBEV Easy | AP|R40: 27.65 -> 30.55 | Improved |
| average top K=2 candidates instead of K=1 in distance-guided NMS (ablation) | UR3D + DG-NMS · KITTI val1 car, AP3D Easy | AP|R40: 19.75 -> 20.20 | Improved |
| add projection-consistency post-optimization (ablation) | UR3D + DG-NMS (K=2) · KITTI val1 car, AP3D Easy | AP|R40: 20.20 -> 22.50 | Improved |
| add fully convolutional cascaded point regression (ablation) | UR3D + post-optimization · KITTI val1 car, AP3D Easy | AP|R40: 22.50 -> 23.24 | Improved |
| add post-optimization (inference time cost) (ablation) | UR3D + DG-NMS (K=2) · KITTI val1 car | inference time (ms): 90 -> 110 | Hurt |
| add cascaded point regression (inference time cost) (ablation) | UR3D + post-optimization · KITTI val1 car | inference time (ms): 110 -> 120 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- M3D-RPN — M3D-RPN: Monocular 3D Region Proposal Network for Object Detection (ICCV 2019) · better on 3 of 7 shared results
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
- RTM3D: Real-time Monocular 3D Detection from Object Keypoints for Autonomous Driving (ECCV 2020 · similarity 0.90)
- Geometry-Based Distance Decomposition for Monocular 3D Object Detection (ICCV 2021 · similarity 0.90)
- MonoUNI: A Unified Vehicle and Infrastructure-side Monocular 3D Object Detection Network with Sufficient Depth Clues (NeurIPS 2023 · similarity 0.90)
- Accurate Monocular 3D Object Detection via Color-Embedded 3D Reconstruction for Autonomous Driving (ICCV 2019 · similarity 0.90)
- Homography Loss for Monocular 3D Object Detection (CVPR 2022 · similarity 0.89)
- M3D-RPN: Monocular 3D Region Proposal Network for Object Detection (ICCV 2019 · 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.
