On June 30, 2026, the Research Centre for Artificial Intelligence in Geomatics (RCAIG) at The Hong Kong Polytechnic University (PolyU) held its monthly academic workshop. Hosted by Prof. Qihao Weng, Chair Professor of Geomatics and Artificial Intelligence and Director of RCAIG, and chaired by Dr. Pir Mohammad, Research Assistant Professor, the workshop served as a vibrant platform to showcase the Centre’s latest research achievements.
The workshop featured highly advanced methodologies and theoretical insights from the Centre's Research Assistant Professor and Postdoctoral Fellows, who addressed complex physical, environmental, and computational challenges. Dr. Pir Mohammad investigated the non-linear relationships between the seasonal and diurnal Surface Urban Heat Island (SUHI) effect and six distinct thematic variable groups. Dr. Chouaib El Hachimi reviewed the current geospatial foundation models (GFMs), proposed a promising framework to enhance physical consistency and interpretability, and presented his evaluation of applying GFMs to informal settlement mapping in Africa. Dr. Baoling Gui re-examined the "trading space for time" paradigm, proposing that researchers must convert spatial equilibrium differences into real-time transient responses through precise temporal corrections, rather than directly equating spatial differences with temporal changes. Dr. Haider Abbas presented projections on how compound humid heat risks evolve across different Shared Socioeconomic Pathways (SSP1-2.6, SSP2-4.5, and SSP5-8.5), demonstrating that future risks are shaped not only by event frequency but also by how they unfold temporally.
The PhD students at RCAIG demonstrated exceptional creativity and technical rigor across a wide range of applications in geospatial AI, ecological modelling, coastal monitoring, and urban sustainability. Ms. Qianbao Hou proposed an innovative, safe, and self-evolving geospatial agent framework for post-disaster resident actionability, which optimises facility-route planning in unseen cities by querying real-time spatial evidence and learning from spatial feedback. Ms. Mengying Cao shared her latest progress in disentangling the complex phenological responses of Arctic vegetation to rapid climate warming. Mr. Hamza Mukhtar demonstrated that delayed spring onset at higher elevations is associated with reduced vegetation productivity, whereas delayed autumn generally enhances productivity in these ecosystems. Mr. Longjie Ye introduced a deep learning framework incorporating a Gaussian loss function to predict mangrove canopy height while systematically quantifying associated prediction uncertainties. Mr. Zhixing Chen presented his research on estimating mobility-related carbon emissions across global cities and identifying the key driving factors behind these emissions. Ms. Wanru He presented her recent progress on the co-amplification of climate and social vulnerabilities, highlighting the methodological and modelling innovations in her assessment framework.
Throughout the workshop, Prof. Qihao Weng actively guided the academic discussions, providing in-depth and constructive feedback on each presentation. This monthly workshop series continues to serve as an invaluable platform for RCAIG researchers to refine their methodologies, exchange innovative ideas, and drive forward-looking research in Geomatics and GeoAI. By fostering such a collaborative and intellectually rigorous environment, RCAIG remains committed to nurturing the next generation of scientific leaders and developing data-driven, space-time solutions to address the world's most pressing environmental and societal challenges.
The workshop featured highly advanced methodologies and theoretical insights from the Centre's Research Assistant Professor and Postdoctoral Fellows, who addressed complex physical, environmental, and computational challenges. Dr. Pir Mohammad investigated the non-linear relationships between the seasonal and diurnal Surface Urban Heat Island (SUHI) effect and six distinct thematic variable groups. Dr. Chouaib El Hachimi reviewed the current geospatial foundation models (GFMs), proposed a promising framework to enhance physical consistency and interpretability, and presented his evaluation of applying GFMs to informal settlement mapping in Africa. Dr. Baoling Gui re-examined the "trading space for time" paradigm, proposing that researchers must convert spatial equilibrium differences into real-time transient responses through precise temporal corrections, rather than directly equating spatial differences with temporal changes. Dr. Haider Abbas presented projections on how compound humid heat risks evolve across different Shared Socioeconomic Pathways (SSP1-2.6, SSP2-4.5, and SSP5-8.5), demonstrating that future risks are shaped not only by event frequency but also by how they unfold temporally.
The PhD students at RCAIG demonstrated exceptional creativity and technical rigor across a wide range of applications in geospatial AI, ecological modelling, coastal monitoring, and urban sustainability. Ms. Qianbao Hou proposed an innovative, safe, and self-evolving geospatial agent framework for post-disaster resident actionability, which optimises facility-route planning in unseen cities by querying real-time spatial evidence and learning from spatial feedback. Ms. Mengying Cao shared her latest progress in disentangling the complex phenological responses of Arctic vegetation to rapid climate warming. Mr. Hamza Mukhtar demonstrated that delayed spring onset at higher elevations is associated with reduced vegetation productivity, whereas delayed autumn generally enhances productivity in these ecosystems. Mr. Longjie Ye introduced a deep learning framework incorporating a Gaussian loss function to predict mangrove canopy height while systematically quantifying associated prediction uncertainties. Mr. Zhixing Chen presented his research on estimating mobility-related carbon emissions across global cities and identifying the key driving factors behind these emissions. Ms. Wanru He presented her recent progress on the co-amplification of climate and social vulnerabilities, highlighting the methodological and modelling innovations in her assessment framework.
Throughout the workshop, Prof. Qihao Weng actively guided the academic discussions, providing in-depth and constructive feedback on each presentation. This monthly workshop series continues to serve as an invaluable platform for RCAIG researchers to refine their methodologies, exchange innovative ideas, and drive forward-looking research in Geomatics and GeoAI. By fostering such a collaborative and intellectually rigorous environment, RCAIG remains committed to nurturing the next generation of scientific leaders and developing data-driven, space-time solutions to address the world's most pressing environmental and societal challenges.