Research map / Architectures and efficiency / CNN design
Convolutional Recurrent Network for Road Boundary Extraction
Justin Liang, Namdar Homayounfar, Wei-Chiu Ma, Shenlong Wang, Raquel Urtasun
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 the structured recurrent polyline model instead of the distance-transform baseline | Road boundary extraction · Road boundaries, test set | F1 at 5 px: 85.0 -> 87.2 | Improved |
| Use the structured recurrent polyline model instead of the distance-transform baseline | Road boundary extraction · Road boundaries, test set | Connectivity: 89.1 -> 99.2 | Improved |
| Add camera input to LiDAR and elevation (ablation) | Convolutional Recurrent Network · Road boundaries, test set | F1 at 5 px: 86.2 -> 87.2 | Improved |
| Add elevation input to LiDAR and camera (ablation) | Convolutional Recurrent Network · Road boundaries, test set | F1 at 5 px: 74.3 -> 87.2 | Improved |
| Add LiDAR and elevation inputs to camera-only input (ablation) | Convolutional Recurrent Network · Road boundaries, test set | F1 at 5 px: 74.3 -> 87.2 | Improved |
| Train with half ground-truth and half predicted deep features rather than only predicted features (ablation) | cSnake · Road boundaries, 2 px threshold | Precision, recall, and F1: 5% better | Improved |
| Train with half ground-truth and half predicted deep features rather than only predicted features (ablation) | cSnake · Road boundaries | Connectivity | No clear effect |
| Predict the full direction map instead of dilated boundary normals (ablation) | cSnake · Road boundaries | Precision, recall, and F1 | Improved |
| Predict the full direction map instead of dilated boundary normals (ablation) | cSnake · Road boundaries | Connectivity: 1% higher | Improved |
Closest papers by meaning
- Hierarchical Recurrent Attention Networks for Structured Online Maps (CVPR 2018 · similarity 0.89)
- In Defense of Pre-Trained ImageNet Architectures for Real-Time Semantic Segmentation of Road-Driving Images (CVPR 2019 · similarity 0.88)
- Efficient Outdoor 3D Point Cloud Semantic Segmentation for Critical Road Objects and Distributed Contexts (ECCV 2020 · similarity 0.88)
- ViiNeuS: Volumetric Initialization for Implicit Neural Surface Reconstruction of Urban Scenes with Limited Image Overlap (CVPR 2025 · similarity 0.85)
- LidaRF: Delving into Lidar for Neural Radiance Field on Street Scenes (CVPR 2024 · similarity 0.85)
- Sketch and Refine: Towards Fast and Accurate Lane Detection (AAAI 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.
