Publications
Domain Adaptation
These works provide an overview of my research on domain shifts in aerial scenes due to seasonal variations between wet and dry seasons in the Global South. The aim of these works is to perform aerial scene analysis and investigate which features Deep Learning models rely on when classifying aerial scenes. Based on the findings, we propose Deep Learning architectures that extract invariant feature representations across wet and dry seasonal domains.
Georgios Voulgaris, Andy Philippides, Jonathan Dolley, Jeremy Reffin, Fiona Marshall, Novi Quadrianto. (2023). “Seasonal Domain Shift in the Global South: Dataset and Deep Features Analysis.” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops. Full text available at CVPR’23 .
Poster:
Georgios Voulgaris, Andy Philippides, Novi Quadrianto. (2022). “Deep Learning Robustness to Domain Shifts During Seasonal Variations.” International Geoscience and Remote Sensing Symposium (IGARSS). Full text available at IGARSS’22 .
Poster:
Presentation:
Semantic Segmentation
These works provide an overview of my research in semantic segmentation of aerial scenes.
Georgios Voulgaris. (2025). Bridging Classical and Modern Computer Vision: PerceptiveNet for Tree Crown Semantic Segmentation.
Georgios Voulgaris, Andy Philippides, Novi Quadrianto. (2023). “Water Physics Aware Semantic Segmentation Through Texture-Biased U-Net Architectures.” International Geoscience and Remote Sensing Symposium (IGARSS). Full text available at IGARSS’23 .
Multimodal Deep Learning Data Fusion
These works provide an overview of my research in using remote sensing data and Deep Learning architectures to detect pollutant plants in China. I used Landsat-8 remote sensors (Thermal Infrared and Operational Land Imager) to create datasets consisting of thermal infrared (bands 10 and 11), Short Wave Infrared (bands 6 and 7), and a geological ratio of the Short-Wave Infrared.
Georgios Voulgaris, Maral Bayaraa, Cristian Rossi. (2025). Detecting Cement Plants with Landsat-8: A Physics-Informed, Multi-Temporal, and Multi-Spectral Deep Learning Fusion Approach.
Cristian Rossi, Nataliya Tkachenko, Maral Bayaraa, Peter Foster, Steven Reece, Kimberly Scott, Georgios Voulgaris, Christophe Christiaen, Matthew McCarten. (2022). “Detection and Characterisation of Pollutant Assets with AI and EO to Prioritise Green Investments: The Geoasset Framework.” International Geoscience and Remote Sensing Symposium (IGARSS). Full text available at IGARSS’22 .
Presentation:
PhD Thesis
This research explores transferable features between domains, proposes Deep Learning architectures that combine salient feature extraction with a wider receptive field and highlights the importance of choosing appropriate feature priors for better model generalisation for domain adaptation and semantic segmentation tasks.