Muhammad Kamran has started his doctorate program in the winter, 2019. His research focuses on object detection, instance segmentation, and building footprint extraction.
Kamran is part of Intelligent Systems for Sustainable Urban Mobility (ISSUM) funded by the Ontario Research Fund, York University, University of Waterloo, and private sector partners (Esri Canada, Fugro Roadware, Miovision, Mircom, Teledyne Optech, Exascale Solutions, Trans-Plan).
Experience
- Jan, 2019–2025: Graduate Research Assistant, GeoICT/ AUSMLab, York University Canada
- Jan, 2019–2025: Teaching Assistant, York University, Canada
- Dec, 2016–Oct, 2018: Lecturer, Sarhad University of Science & Information Technology, Pakistan.
- Feb, 2016–Oct, 2016: Solution Provisioning Engineer, IT Consulting & Services, Islamabad, Pakistan.
Publications (5)
2025
- Kamran, M., Sheikholeslami, M. M., Sohn, G. (2025). PolyAttractNet: Graph-Based Polygonal Segmentation of Building Footprints Using Attraction Field Maps. ISPRS Archives, XLVIII-G-2025, 765–771.
- Kamran, M., Sheikholeslami, M. M., Wichmann, A., Sohn, G. (2025). Transformer Based Building Boundary Reconstruction using Attraction Field Maps. arXiv:2507.17038. (Preprint)
2024
- Sheikholeslami, M. M., Kamran, M., Wichmann, A., Sohn, G. (2024). CornerRegNet: Building Segmentation from Overhead Imagery Using Oriented Corners in Deep Networks. IGARSS 2024, 4642–4647.
- Sheikholeslami, M. M., Kamran, M., Wichmann, A., Sohn, G. (2024). Enhancing Polygonal Building Segmentation via Oriented Corners. arXiv:2407.12256. (Preprint)
2020
- Zhao, K., Kamran, M., and Sohn, G. (2020). Boundary Regularized Building Footprint Extraction from Satellite Images Using Deep Neural Network. arXiv preprint arXiv:2006.13176.