Research map / 3D vision / 3D generation
KITTI-360: A Novel Dataset and Benchmarks for Urban Scene Understanding in 2D and 3D
Yiyi Liao, Jun Xie, Andreas Geiger
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 3D points terms to local appearance (ablation) | LA label transfer model · KITTI-360 label transfer (static) | semantic mIoU: 67.6 -> 70.4 | Improved |
| add 3D points terms to local appearance (ablation) | LA label transfer model · KITTI-360 label transfer (static) | instance mIoU: 72.3 -> 72.9 | Improved |
| add 2D pairwise constraints to local appearance (ablation) | LA label transfer model · KITTI-360 label transfer (static) | semantic mIoU: 67.6 -> 66.6 | Hurt |
| add 3D primitive constraints to LA+PW (ablation) | LA+PW · KITTI-360 label transfer (static) | semantic mIoU: 66.6 -> 76.8 | Improved |
| add 3D points terms to LA+PW+CO (ablation) | LA+PW+CO · KITTI-360 label transfer (static) | semantic mIoU: 76.8 -> 78.2 | Improved |
| add 3D pairwise constraints (full model) (ablation) | LA+PW+CO+3D · KITTI-360 label transfer (static) | semantic mIoU: 78.2 -> 81.2 | Improved |
| add 3D pairwise constraints (full model) (ablation) | LA+PW+CO+3D · KITTI-360 label transfer (static) | instance mIoU: 83.6 -> 83.7 | No clear effect |
| predict only the 70% most certain pixels | Full Model · KITTI-360 label transfer (static) | semantic mIoU: 81.2 -> 94.3 | Improved |
| predict only the 70% most certain pixels | Full Model · KITTI-360 label transfer (static) | instance mIoU: 83.7 -> 92.7 | Improved |
| use 3D points projection instead of 3D primitives before graph cuts | 3D to 2D baselines · KITTI-360 label transfer (static) | semantic mIoU: 49.4 -> 72.6 | Improved |
| proposed method vs best 2D label transfer baseline | Label transfer · KITTI-360 label transfer (dynamic) | mIoU: 37.2 -> 63.5 | Improved |
| use deeper ResNet-101 backbone instead of ResNet-50 | Mask R-CNN · KITTI-360 2D instance segmentation | AP: 19.5 -> 20.9 | Improved |
Closest papers by meaning
- The Road Ahead in Autonomous Driving: The KITScenes Multimodal Dataset (NeurIPS 2026 · similarity 0.88)
- MAPLM: A Real-World Large-Scale Vision-Language Benchmark for Map and Traffic Scene Understanding (CVPR 2024 · similarity 0.86)
- DriveSpatial: A Benchmark for Spatiotemporal Intelligence in VLMs for Autonomous Driving (NeurIPS 2026 · similarity 0.85)
- UrbanIng-V2X: A Large-Scale Multi-Vehicle, Multi-Infrastructure Dataset Across Multiple Intersections for Cooperative Perception (NeurIPS 2025 · similarity 0.85)
- RoboSpatial: Teaching Spatial Understanding to 2D and 3D Vision-Language Models for Robotics (CVPR 2025 · similarity 0.85)
- RoadSceneBench: A Lightweight Benchmark for Mid-Level Road Scene Understanding (CVPR 2026 · similarity 0.85)
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.
