Holistically-Nested Structure-Aware Graph Neural Network for Road Extraction
This work addresses a domain-specific problem in remote sensing for applications like mapping and navigation, offering an incremental improvement over prior CNN-based approaches.
The paper tackles the problem of poor delineation and connectivity in road extraction from satellite images by introducing a multi-task graph neural network that simultaneously detects road regions and borders, resulting in improved accuracy and border delineation compared to existing methods.
Convolutional neural networks (CNN) have made significant advances in detecting roads from satellite images. However, existing CNN approaches are generally repurposed semantic segmentation architectures and suffer from the poor delineation of long and curved regions. Lack of overall road topology and structure information further deteriorates their performance on challenging remote sensing images. This paper presents a novel multi-task graph neural network (GNN) which simultaneously detects both road regions and road borders; the inter-play between these two tasks unlocks superior performance from two perspectives: (1) the hierarchically detected road borders enable the network to capture and encode holistic road structure to enhance road connectivity (2) identifying the intrinsic correlation of semantic landcover regions mitigates the difficulty in recognizing roads cluttered by regions with similar appearance. Experiments on challenging dataset demonstrate that the proposed architecture can improve the road border delineation and road extraction accuracy compared with the existing methods.