DeH4R: A Decoupled and Hybrid Method for Road Network Graph Extraction

The automated extraction of complete and precise road network graphs from remote sensing imagery remains a critical challenge in geospatial computer vision. Segmentationbased approaches, while effective in pixel-level recognition, struggle to maintain topology fidelity after vectorization post-processing. Graph-growing methods build more topologically faithful graphs but suffer from computationally prohibitive iterative ROI cropping. Graph-generating methods first predict global static candidate road network vertices, and then infer possible edges between vertices. They achieve fast topology-aware inference, but limits the dynamic insertion of vertices. To address these challenges, we propose DeH4R, a novel hybrid model that combines graph-generating efficiency and graph-growing dynamics. This is achieved by decoupling the task into candidate vertex detection, adjacent vertex prediction, initial graph construction, and graph expansion. This architectural innovation enables dynamic vertex (edge) insertions while retaining fast inference speed and enhancing both topology fidelity and spatial consistency. Moreover, it is simple and straightforward to implement. Comprehensive evaluations on the CityScale and SpaceNet benchmarks demonstrate state-of-the-art (SOTA) performance. DeH4R outperforms the prior SOTA graph-growing method RNGDet++ by 4.66 APLS and 10.18 IoU on CityScale, while being approximately 10× faster. The code is publicly available at https://github.com/7777777FAN/DeH4R.