Tavis Shore

Profile

I'm a PhD researcher in Computer Vision at the University of Surrey currently working on cross-view geo-localisation. I work with Dr. Oscar Mendez and Dr. Simon Hadfield in the Centre for Vision, Speech and Signal Processing (CVSSP). My research focuses on developing novel deep learning approaches for precise localisation using ground-level and satellite imagery. My work aims to enhance the real-world applicability of vision-based localisation systems. I've developed several innovative approaches including PEnG, which combines coarse geo-localisation with precise pose estimation, and SpaGBOL, which leverages graph neural networks to model spatial relationships between locations. I'm particularly interested in making these systems work with realistic constraints like limited field-of-view cameras and urban environments.

Publications

PEnG: Pose-Enhanced Geo-Localisation

PEnG: Pose-Enhanced Geo-Localisation

Tavis Shore, Oscar Mendez, Simon Hadfield

arXiv.org 2024

SpaGBOL: Spatial-Graph-Based Orientated Localisation

SpaGBOL: Spatial-Graph-Based Orientated Localisation

Tavis Shore, Oscar Mendez, Simon Hadfield

IEEE/CVF Winter Conference on Applications of Computer Vision 2025

BEV-CV: Birds-Eye-View Transform for Cross-View Geo-Localisation

BEV-CV: Birds-Eye-View Transform for Cross-View Geo-Localisation

Tavis Shore, Simon Hadfield, Oscar Mendez

IEEE/RJS International Conference on Intelligent Robots and Systems 2023

Constrained Machine Learning for LoRa Gateway Location Optimisation

Constrained Machine Learning for LoRa Gateway Location Optimisation

A. Abubakar, Tavis Shore, Nishanth R. Sastry

Asian Internet Engineering Conference 2022