Publications

FieldMapper: Automated Agricultural Field Delineation via Vision Foundation Model and Machinery Trajectories

Shuliang Wang, Haoyu Han, Sijie Ruan*, Ziyu Wang, Ziqiang Yuan, Yeting Zhang, Hanning Yuan and Caicong Wu
in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2025

FieldMapper achieves high-precision agricultural field boundary extraction by fusing remote sensing imagery and agricultural machinery trajectory data, leveraging the instance segmentation capabilities of SAM, without the need for dense manual annotation.

Recommended citation: S. Wang et al., "FieldMapper: Automated Agricultural Field Delineation via Vision Foundation Model and Machinery Trajectories," in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 18, pp. 26054-26067, 2025, doi: 10.1109/JSTARS.2025.3618149.
URL: https://ieeexplore.ieee.org/document/11193714

Spatial Hierarchical Meta-Learning for Single-Point Map Matching

Sijie Ruan, Yiqing Zou, Qianyu Yang, Haoyu Han, Yeting Zhang, Ziqiang Yuan, Hanning Yuan and Shuliang Wang.
in Proceedings of the 31th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2024

SHSMM leverages gradient-based meta-learning to unify region-specific heterogeneity and cross-task commonality for single-point map matching in location-based services.

Recommended citation: Sijie Ruan, Yiqing Zou, Qianyu Yang, Haoyu Han, Yeting Zhang, Ziqiang Yuan, Hanning Yuan and Shuliang Wang. 2025. Spatial Hierarchical Meta-Learning for Single-Point Map Matching. In Proceedings of the 31st ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD'25). Association for Computing Machinery, New York, NY, USA, 2455–2465.
URL: https://doi.org/10.1145/3711896.3737133

DiffCrime: A Multimodal Conditional Diffusion Model for Crime Risk Map Inference

Shuliang Wang, Xinyu Pan, Sijie Ruan*, Haoyu Han, Ziyu Wang, Hanning Yuan, Jiabao Zhu and Qi Li
in Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2024

DiffCrime utilizes a multimodal diffusion model to integrate historical cases, satellite imagery, and map data, and accurately generates urban crime risk maps through HamNet. The Root Mean Squared Error (RMSE) of the two real-world datasets is reduced by 43% and 31%, respectively.

Recommended citation: Shuliang Wang, Xinyu Pan, Sijie Ruan, Haoyu Han, Ziyu Wang, Hanning Yuan, Jiabao Zhu, and Qi Li. 2024. DiffCrime: A Multimodal Conditional Diffusion Model for Crime Risk Map Inference. In Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD '24). Association for Computing Machinery, New York, NY, USA, 3212–3221.
URL: https://doi.org/10.1145/3637528.3671843

DelvMap: Completing Residential Roads in Maps Based on Couriers’ Trajectories and Satellite Imagery

Shuliang Wang, Ziyu Wang, Sijie Ruan*, Haoyu Han, Keqin Xiong, Hannning Yuan, Ziqiang Yuan, Guoqing Li, Jie Bao and Yu Zheng
in IEEE Transactions on Geoscience and Remote Sensing, 2024

This paper is about residential-level fine-grained digital map for last-mile delivery. Based on couriers’ trajectories and satellite images, this model has a better performance in real-world dataset.

Recommended citation: S. Wang et al., "DelvMap: Completing Residential Roads in Maps Based on Couriers’ Trajectories and Satellite Imagery," in IEEE Transactions on Geoscience and Remote Sensing, doi: 10.1109/TGRS.2024.3365833.
URL: https://ieeexplore.ieee.org/document/10436072