Beyond Endpoints: Path-Centric Reasoning for Vectorized Off-Road Network Extraction
- URL: http://arxiv.org/abs/2512.10416v2
- Date: Fri, 12 Dec 2025 05:10:11 GMT
- Title: Beyond Endpoints: Path-Centric Reasoning for Vectorized Off-Road Network Extraction
- Authors: Wenfei Guan, Jilin Mei, Tong Shen, Xumin Wu, Shuo Wang, Cheng Min, Yu Hu,
- Abstract summary: We release WildRoad, a global off-road road network dataset constructed efficiently with a dedicated interactive annotation tool.<n>We introduce MaGRoad, a path-centric framework that aggregates multi-scale visual evidence along candidate paths to infer connectivity robustly.<n>MaGRoad achieves state-of-the-art performance on our challenging WildRoad benchmark while generalizing well to urban datasets.
- Score: 9.833728353188132
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Deep learning has advanced vectorized road extraction in urban settings, yet off-road environments remain underexplored and challenging. A significant domain gap causes advanced models to fail in wild terrains due to two key issues: lack of large-scale vectorized datasets and structural weakness in prevailing methods. Models such as SAM-Road employ a node-centric paradigm that reasons at sparse endpoints, making them fragile to occlusions and ambiguous junctions in off-road scenes, leading to topological errors. This work addresses these limitations in two complementary ways. First, we release WildRoad, a global off-road road network dataset constructed efficiently with a dedicated interactive annotation tool tailored for road-network labeling. Second, we introduce MaGRoad (Mask-aware Geodesic Road network extractor), a path-centric framework that aggregates multi-scale visual evidence along candidate paths to infer connectivity robustly. Extensive experiments show that MaGRoad achieves state-of-the-art performance on our challenging WildRoad benchmark while generalizing well to urban datasets. A streamlined pipeline also yields roughly 2.5x faster inference, improving practical applicability. Together, the dataset and path-centric paradigm provide a stronger foundation for mapping roads in the wild. We release both the dataset and code at https://github.com/xiaofei-guan/MaGRoad.
Related papers
- Towards Satellite Image Road Graph Extraction: A Global-Scale Dataset and A Novel Method [42.3609654615897]
We collect a global-scale satellite road graph extraction dataset, i.e. Global-Scale dataset.
We develop a novel road graph extraction model, i.e. SAM-Road++, which adopts a node-guided resampling method.
Experiments demonstrate the validity of the collected Global-Scale dataset and the proposed SAM-Road++ method.
arXiv Detail & Related papers (2024-11-23T10:26:07Z) - Reverse Refinement Network for Narrow Rural Road Detection in High-Resolution Satellite Imagery [6.582832730721428]
R2-Net is proposed to extract narrow rural roads, enhancing their connectivity and distinctiveness from the background.
In experiments, we compare R2-Net with several state-of-the-art methods using the DeepGlobe road extraction dataset and the WHU-RuR+ global large-scale rural road dataset.
The results show that the proposed R2-Net has significant performance advantages for large-scale rural road mapping applications.
arXiv Detail & Related papers (2024-10-14T11:23:47Z) - Holistically-Nested Structure-Aware Graph Neural Network for Road Extraction [6.537257686406911]
This paper presents a novel graph neural network (GNN) which simultaneously detects both road regions and road borders.
Experiments on challenging dataset demonstrate that the proposed architecture can improve the road border delineation and road extraction accuracy.
arXiv Detail & Related papers (2024-07-02T20:11:09Z) - Translating Images to Road Network: A Sequence-to-Sequence Perspective [30.616209208000097]
Road network is essential for the generation of high-definition maps.<n>Existing methods struggle to merge the two types of data domains effectively.<n>We propose a unified representation of both types of data domain by projecting both Euclidean and non-Euclidean data into an integer series called RoadNet Sequence.
arXiv Detail & Related papers (2024-02-13T04:12:41Z) - Leveraging Driver Field-of-View for Multimodal Ego-Trajectory Prediction [69.29802752614677]
RouteFormer is a novel ego-trajectory prediction network combining GPS data, environmental context, and the driver's field-of-view.<n>To tackle data scarcity and enhance diversity, we introduce GEM, a dataset of urban driving scenarios enriched with synchronized driver field-of-view and gaze data.
arXiv Detail & Related papers (2023-12-13T23:06:30Z) - Graph-based Topology Reasoning for Driving Scenes [102.35885039110057]
We present TopoNet, the first end-to-end framework capable of abstracting traffic knowledge beyond conventional perception tasks.
We evaluate TopoNet on the challenging scene understanding benchmark, OpenLane-V2.
arXiv Detail & Related papers (2023-04-11T15:23:29Z) - PaRK-Detect: Towards Efficient Multi-Task Satellite Imagery Road
Extraction via Patch-Wise Keypoints Detection [12.145321599949236]
We propose a new scheme for multi-task satellite imagery road extraction, Patch-wise Road Keypoints Detection (PaRK-Detect)
Our framework predicts the position of patch-wise road keypoints and the adjacent relationships between them to construct road graphs in a single pass.
We evaluate our approach against the existing state-of-the-art methods on DeepGlobe, Massachusetts Roads, and RoadTracer datasets and achieve competitive or better results.
arXiv Detail & Related papers (2023-02-26T08:26:26Z) - Road Network Guided Fine-Grained Urban Traffic Flow Inference [108.64631590347352]
Accurate inference of fine-grained traffic flow from coarse-grained one is an emerging yet crucial problem.
We propose a novel Road-Aware Traffic Flow Magnifier (RATFM) that exploits the prior knowledge of road networks.
Our method can generate high-quality fine-grained traffic flow maps.
arXiv Detail & Related papers (2021-09-29T07:51:49Z) - SPIN Road Mapper: Extracting Roads from Aerial Images via Spatial and
Interaction Space Graph Reasoning for Autonomous Driving [64.10636296274168]
Road extraction is an essential step in building autonomous navigation systems.
Using just convolution neural networks (ConvNets) for this problem is not effective as it is inefficient at capturing distant dependencies between road segments in the image.
We propose a Spatial and Interaction Space Graph Reasoning (SPIN) module which when plugged into a ConvNet performs reasoning over graphs constructed on spatial and interaction spaces projected from the feature maps.
arXiv Detail & Related papers (2021-09-16T03:52:17Z) - DAGMapper: Learning to Map by Discovering Lane Topology [84.12949740822117]
We focus on drawing the lane boundaries of complex highways with many lanes that contain topology changes due to forks and merges.
We formulate the problem as inference in a directed acyclic graphical model (DAG), where the nodes of the graph encode geometric and topological properties of the local regions of the lane boundaries.
We show the effectiveness of our approach on two major North American Highways in two different states and show high precision and recall as well as 89% correct topology.
arXiv Detail & Related papers (2020-12-22T21:58:57Z) - Convolutional Recurrent Network for Road Boundary Extraction [99.55522995570063]
We tackle the problem of drivable road boundary extraction from LiDAR and camera imagery.
We design a structured model where a fully convolutional network obtains deep features encoding the location and direction of road boundaries.
We showcase the effectiveness of our method on a large North American city where we obtain perfect topology of road boundaries 99.3% of the time.
arXiv Detail & Related papers (2020-12-21T18:59:12Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.