Research map / Detection and segmentation / 3D detection and driving
"The Pedestrian Next to the Lamppost" Adaptive Object Graphs for Better Instantaneous Mapping
Avishkar Saha, Oscar Mendez, Chris Russell, Richard Bowden
Check 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 |
|---|---|---|---|
| Use object-graph reasoning instead of the strongest listed BEV estimator | BEV estimation · nuScenes val | Objects Mean IoU (%): 18.1 -> 27.1 | Improved |
| Use object-graph reasoning instead of the strongest listed BEV estimator | BEV estimation · Argoverse val | Mean IoU (%): 22.6 -> 30.3 | Improved |
| Add edge-to-node communication to node-to-node propagation (ablation) | Graph propagation on nuScenes · nuScenes | Objects Mean IoU (%): 20.0 -> 21.1 | Improved |
| Add edge-to-edge propagation and edge supervision (ablation) | Graph propagation on nuScenes · nuScenes | Objects Mean IoU (%): 21.1 -> 25.9 | Improved |
| Add node-to-edge communication (ablation) | Graph propagation on nuScenes · nuScenes | Objects Mean IoU (%): 25.9 -> 27.1 | Improved |
| Add scanline features to appearance features (ablation) | Graph node and edge features · nuScenes | Objects Mean IoU (%): 22.5 -> 22.9 | No clear effect |
| Add geometry features to appearance features (ablation) | Graph node and edge features · nuScenes | Objects Mean IoU (%): 22.5 -> 26.1 | Improved |
| Add positional information to appearance, scanline, and geometry features (ablation) | Graph node and edge features · nuScenes | Objects Mean IoU (%): 26.2 -> 27.1 | Improved |
| Increase graph node degree from 3 to 5 (ablation) | Graph construction · nuScenes | Objects Mean IoU (%): 27.1 -> 21.1 | Hurt |
| Increase graph node degree from 3 to 10 (ablation) | Graph construction · nuScenes | Objects Mean IoU (%): 27.1 -> 14.2 | Hurt |
| Increase graph node degree from 3 to 15 (ablation) | Graph construction · nuScenes | Objects Mean IoU (%): 27.1 -> 11.56 | Hurt |
| Increase graph node degree from 3 to 20+ (ablation) | Graph construction · nuScenes | Objects Mean IoU (%): 27.1 -> 10.3 | Hurt |
Closest papers by meaning
- Improving Bird's Eye View Semantic Segmentation by Task Decomposition (CVPR 2024 · similarity 0.90)
- BEV-CAR: Enhancing Monocular Bird's Eye View Segmentation with Context-Aware Rasterization (CVPR 2026 · similarity 0.89)
- Structured Bird's-Eye-View Traffic Scene Understanding From Onboard Images (ICCV 2021 · similarity 0.89)
- Map It Anywhere: Empowering BEV Map Prediction using Large-scale Public Datasets (NeurIPS 2024 · similarity 0.87)
- A Black-Box Evaluation Framework for Semantic Robustness in Bird’s Eye View Detection (AAAI 2025 · similarity 0.86)
- UniMODE: Unified Monocular 3D Object Detection (CVPR 2024 · 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.
