Research map / 3D vision / Human pose estimation
Robust 2D/3D Vehicle Parsing in Arbitrary Camera Views for CVIS
Hui Miao, Feixiang Lu, Zongdai Liu, Liangjun Zhang, Dinesh Manocha, Bin Zhou
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 |
|---|---|---|---|
| train on our part-assisted synthesized data instead of existing front-view dataset | 2D/3D vehicle parsing network · CVIS test set, 2D detection | mAP: 49.8 -> 53.6 | Improved |
| train on our synthesized data instead of rendering data | 2D/3D vehicle parsing network · CVIS test set, 2D detection | mAP: 46.0 -> 53.6 | Improved |
| train on our synthesized data instead of GAN-based data | 2D/3D vehicle parsing network · CVIS test set, instance segmentation | mAP: 37.4 -> 52.4 | Improved |
| train on our synthesized data instead of existing dataset | 2D/3D vehicle parsing network · CVIS test set, 6-DoF pose | Abs-mean: 22.4 -> 25.3 | Improved |
| use part-based texture inpainting instead of KNN filling (ablation) | 2D/3D vehicle parsing network · CVIS test set, 2D detection | mAP: 57.5 -> 59.7 | Improved |
| use part-based texture inpainting instead of KNN filling (ablation) | 2D/3D vehicle parsing network · CVIS test set, instance segmentation | mAP: 51.1 -> 54.6 | Improved |
| use part-based texture inpainting instead of generic image inpainting (ablation) | 2D/3D vehicle parsing network · CVIS test set, instance segmentation | mAP: 47.4 -> 54.6 | Improved |
| add vehicle shadows to synthesized data (ablation) | 2D/3D vehicle parsing network · CVIS test set, 2D detection | mAP: 58.3 -> 59.7 | Improved |
| add vehicle shadows to synthesized data (ablation) | 2D/3D vehicle parsing network · CVIS test set, 6-DoF pose | Abs-mean: 30.9 -> 33.0 | Improved |
| add vehicle shadows to synthesized data (ablation) | 2D/3D vehicle parsing network · CVIS test set, instance segmentation | mAP: 53.1 -> 54.6 | Improved |
| use dense-mapping parsing network (ours) instead of DensePose | pose estimation · CVIS test set, 6-DoF pose | A3DP-Abs mean: 28.7 -> 38.6 | Improved |
| mix our synthesized CVIS data with KITTI training data | Mask-RCNN · KITTI 2D detection | mAP: 43.7 -> 45.1 | Improved |
Closest papers by meaning
- VQA-Diff: Exploiting VQA and Diffusion for Zero-Shot Image-to-3D Vehicle Asset Generation in Autonomous Driving (ECCV 2024 · similarity 0.86)
- DI-V2X: Learning Domain-Invariant Representation for Vehicle-Infrastructure Collaborative 3D Object Detection (AAAI 2024 · similarity 0.85)
- Structural Information Guided Multimodal Pre-training for Vehicle-Centric Perception (AAAI 2024 · similarity 0.84)
- Vanishing-Point-Guided Video Semantic Segmentation of Driving Scenes (CVPR 2024 · similarity 0.84)
- UrbanIng-V2X: A Large-Scale Multi-Vehicle, Multi-Infrastructure Dataset Across Multiple Intersections for Cooperative Perception (NeurIPS 2025 · similarity 0.84)
- VR-Drive: Viewpoint-Robust End-to-End Driving with Feed-Forward 3D Gaussian Splatting (NeurIPS 2025 · similarity 0.84)
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.
