Research map / Detection and segmentation / 3D detection and driving
A Continuous Occlusion Model for Road Scene Understanding
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 |
|---|---|---|---|
| Use continuous 3D occlusion reasoning for point-track association instead of bounding-box heuristics | Point-to-object association · KITTI raw, dynamic & occluded point tracks | mean association error: 21.3 -> 13.2 | Helped |
| Use continuous 3D occlusion reasoning for point-track association instead of motion segmentation | Point-to-object association · KITTI raw, dynamic & occluded point tracks | mean association error: 30.1 -> 13.2; 30.9 -> 13.2 | Helped |
| Use continuous 3D occlusion reasoning for point-track association instead of bounding-box heuristics | Point-to-object association · KITTI raw, occluded point tracks | mean association error: 19.8 -> 15.7 | Helped |
| Use continuous 3D occlusion reasoning for point-track association instead of motion segmentation | Point-to-object association · KITTI raw, occluded point tracks | mean association error: 39.5 -> 15.7; 37.8 -> 15.7 | Helped |
| Use continuous 3D occlusion reasoning for point-track association instead of bounding-box heuristics | Point-to-object association · KITTI raw, dynamic point tracks | mean association error: 11.4 -> 6.6 | Helped |
| Use continuous 3D occlusion reasoning for point-track association instead of motion segmentation | Point-to-object association · KITTI raw, dynamic point tracks | mean association error: 15.3 -> 6.6; 17.7 -> 6.6 | Helped |
| Use continuous 3D occlusion reasoning for point-track association instead of bounding-box heuristics | Point-to-object association · KITTI raw, all point tracks | mean association error: 12.6 -> 8.6 | Helped |
| Use continuous 3D occlusion reasoning for point-track association instead of motion segmentation | Point-to-object association · KITTI raw, all point tracks | mean association error: 21.5 -> 8.6; 21.9 -> 8.6 | Helped |
| Add occlusion-aware point-track and detection energies | 3D object localization · KITTI raw, occluded tracks | translation error (m/car): 3.95 -> 3.24 | Helped |
| Add occlusion-aware point-track and detection energies | 3D object localization · KITTI raw, occluded tracks | dimension error (m/car): 1.72 -> 2.16 | Hurt |
| Add occlusion modeling to the detection energy (ablation) | 3D object localization · KITTI raw, occluded tracks | translation error (m/car): 4.81 -> 3.24 | Helped |
| Add occlusion modeling to the detection energy (ablation) | 3D object localization · KITTI raw, occluded tracks | dimension error (m/car): 1.59 -> 2.16 | Hurt |
Closest papers by meaning
- Towards Flexible 3D Perception: Object-Centric Occupancy Completion Augments 3D Object Detection (NeurIPS 2024 · similarity 0.87)
- CorrBEV: Multi-View 3D Object Detection by Correlation Learning with Multi-modal Prototypes (CVPR 2025 · similarity 0.86)
- PanoOcc: Unified Occupancy Representation for Camera-based 3D Panoptic Segmentation (CVPR 2024 · similarity 0.86)
- WALT3D: Generating Realistic Training Data from Time-Lapse Imagery for Reconstructing Dynamic Objects Under Occlusion (CVPR 2024 · similarity 0.85)
- CSV-Occ: Fusing Multi-frame Alignment for Occupancy Prediction with Temporal Cross State Space Model and Central Voting Mechanism (ICML 2025 · similarity 0.85)
- 3D StreetUnveiler with Semantic-aware 2DGS - a simple baseline (ICLR 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.