Rhodri Guerrier

I am a computer vision PhD student at the University of Bristol, supervised by Professor Dima Damen. My research focuses on point tracking and 3D reconstruction, with an emphasis on applying this to egocentric video.

Email  /  Scholar  /  Github

Research

Simple fine-tuning of MASt3R foundation model produces competitive point tracking results to state-of-the-art.

A highly detailed kitchen-based egocentric dataset, manually annotated with ground truth labels covering: recipe steps, fine-grained actions, ingredients with nutritional values, moving objects, and audio annotations. All annotations are grounded in 3D through a digital twin of each environment.

A benchmark for point tracking in egocentric visions which highlights the difficulty current models face performing under the scenarios such as motion blur and frequent re-identification.