Charalambos Poullis

Latest — Mar 26, 2026

Data Leakage Detection and De-duplication in Large Scale Geospatial Image Datasets

The paper “Data Leakage Detection and De-duplication in Large Scale Geospatial Image Datasets” by Yeshwanth Kumar Adimoolam, Charalambos Poullis, and Melinos Averkiou will appear in IEEE/CVF Computer Vision and Pattern Recognition (CVPR), 2026. TL;DR: The AICrowd Mapping Challenge dataset is riddled with problems - ~90% duplicate training images

More issues

A novel AI-driven approach to redirected walking in virtual reality that eliminates the need for eye-tracking hardware.

Our patent "Methods and Systems for Real-Time Saccade Prediction" has been granted. TL;DR: A machine learning system that predicts natural saccadic eye movements in real-time, enabling redirected walking in virtual environments by leveraging inattentional blindness—without requiring expensive eye-tracking equipment or artificially triggering major saccades in users.

Extreme Views: 3DGS Filter for Novel View Synthesis from Out‑of‑Distribution Camera Poses

The paper “Extreme Views: 3DGS Filter for Novel View Synthesis from Out‑of‑Distribution Camera Poses” by Damian Bowness and Charalambos Poullis will appear in the 20th International Symposium on Visual Computing (ISVC), 2025. TL;DR: A plug‑in, real‑time filter for 3D Gaussian Splatting that performs a rotation‑

Fast Self-Supervised Depth and Mask Aware Association for Multi-Object Tracking

The paper “Fast Self-Supervised Depth and Mask Aware Association for Multi-Object Tracking” by Milad Khanchi, Maria Amer, and Charalambos Poullis has been accepted for publication at British Machine Vision Conference (BMVC) 2025. TL;DR: SelfTrEncMOT is a novel multi-object tracking framework that integrates zero-shot monocular depth estimation and promptable segmentation

Depth-Aware Scoring and Hierarchical Alignment for Multiple Object Tracking

The paper 'Depth-Aware Scoring and Hierarchical Alignment for Multiple Object Tracking' by Milad Khanchi, Maria Amer, and Charalambos Poullis has been accepted for publication in IEEE International Conference on Image Processing (ICIP) 2025. TL;DR: DepthMOT is a cutting-edge multiple object tracking framework that enhances object association by

DSV-LFS: Unifying LLM-Driven Semantic Cues with Visual Features for Robust Few-Shot Segmentation

The paper 'DSV-LFS: Unifying LLM-Driven Semantic Cues with Visual Features for Robust Few-Shot Segmentation' by Amin Karimi and Charalambos Poullis has been accepted for publication in IEEE/CVF Computer Vision and Pattern Recognition (CVPR) 2025. TL;DR: The paper introduces DSV-LFS, a framework that boosts few-shot semantic segmentation

Pix2Poly: A Sequence Prediction Method for End-to-end Polygonal Building Footprint Extraction from Remote Sensing Imagery

The paper 'Pix2Poly: A Sequence Prediction Method for End-to-end Polygonal Building Footprint Extraction from Remote Sensing Imagery' by Yeshwanth Kumar Adimoolam, Charalambos Poullis, and Melinos Averkiou has been accepted for publication in IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) 2025. TL;DR: This work introduces

From data to action in flood forecasting leveraging graph neural networks and digital twin visualization

The paper "From data to action in flood forecasting leveraging graph neural networks and digital twin visualization" by Naghmeh Shafiee Roudbari, Shubham Rajeev Punekar, Zachary Patterson, Ursula Eicker, and Charalambos Poullis has been accepted for publication in Scientific Reports. Le Devoir published an article about the work in

HydroVision: LiDAR-Guided Hydrometric Prediction with Vision Transformers and Hybrid Graph Learning

The paper "HydroVision: LiDAR-Guided Hydrometric Prediction with Vision Transformers and Hybrid Graph Learning" by Naghmeh Shafiee Roudbari, Ursula Eicker, Charalambos Poullis, Zachary Patterson has been accepted for publication in 19th International Symposium on Visual Computing (ISVC), 2024. TL:DR: A hybrid graph learning model using LiDAR terrain elevation

Transductive meta‑learning with enhanced feature ensemble for few‑shot semantic segmentation

The paper "Transductive Meta-Learning with Enhanced Feature Ensemble for Few-shot Semantic Segmentation" by Amin Karimi and Charalambos Poullis has been accepted for publication in Scientific Reports. TL;DR: We propose two-pass end-to-end method for few-shot semantic segmentation. The approach leverages an ensemble of visual features learned from pretrained