Abstract | ||
---|---|---|
Road network extraction from satellite images often produce fragmented road segments leading to road maps unfit for real applications. Pixel-wise classification fails to predict topologically correct and connected road masks due to the absence of connectivity supervision and difficulty in enforcing topological constraints. In this paper; we propose a connectivity task called Orientation Learning, motivated by the human behavior of annotating roads by tracing it at a specific orientation. We also develop a stacked multibranch convolutional module to effectively utilize the mutual information between orientation learning and segmentation tasks. These contributions ensure that the model predicts topologically correct and connected road masks. We also propose Connectivity Refinement approach to ffirther enhance the estimated road networks. The refinement model is pre-trained to connect and refine the corrupted groundtruth masks and later fine-tuned to enhance the predicted road masks. We demonstrate the advantages of our approach on two diverse road extraction datasets SpaceNet [36] and DeepGlobe [11]. Our approach improves over the state-of-the-art techniques by 9% and 7.5% in road topology metric on SpaceNet and DeepGlobe, respectively. |
Year | DOI | Venue |
---|---|---|
2019 | 10.1109/CVPR.2019.01063 | 2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2019) |
Field | DocType | ISSN |
Computer vision,Pattern recognition,Computer science,Segmentation,Artificial intelligence | Conference | 1063-6919 |
Citations | PageRank | References |
8 | 0.46 | 0 |
Authors | ||
6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Anil Batra | 1 | 9 | 0.81 |
Suriya Singh | 2 | 23 | 3.37 |
Guan Pang | 3 | 42 | 5.81 |
Saikat Basu | 4 | 85 | 7.05 |
C. V. Jawahar | 5 | 1700 | 148.58 |
Manohar Paluri | 6 | 1237 | 56.52 |