Bio

I am a third-year PhD student at The University of York, supervised by Prof. William Smith. My research focuses on computer vision-based scene understanding, emphasising the integration of non-visual features and semantic or temporal cues. I aim to minimise reliance on task-specific data, reducing the need for extensive dataset curation, to enhance the flexibility and adaptability of computer vision systems for diverse applications. I am particularly interested in test-time training, large foundation models, and diffusion-based techniques, primarily for addressing segmentation and tracking tasks in video data.

Timeline


  1. University of York
    PhD Student
    2021 -

  2. Spectral Compute
    GPU Software Engineer
    2023 -

  3. Owl & Lark
    Lead AI Engineer
    2024 -

  4. Satis AI
    Lead Computer Vision Engineer
    2022 - 2023

  5. Visio Impulse
    Computer Vision Scientist
    2020 - 2021

  6. HayBeeSee
    Robotics Engineer
    2018 - 2020

Publications


Track Anything Behind Everything: Zero-Shot Amodal Video Object Segmentation

[Project Page] [Paper] [Code] [Dataset - TABE-51] [Code - Dataset Generation]

Sample Image
If At First You Don't Succeed: Test Time Re-ranking for Zero-shot, Cross-domain Retrieval

[Project Page] [Paper] [Code]

Education


  1. University of York
    PhD in Computer Vision and Machine Learning
    2021 -

  2. University of York
    MSc Intelligent Robotics
    Distinction
    2017 - 2018

  3. Anglia Ruskin University, Cambridge
    BSc (Hons), Audio and Music Technology
    1st Class + Richer Sounds Award
    2014 - 2017

Credit to James Gardner and Yao-Chih Lee for inspiring elements of this page design.