Kieran Saunders

PhD Candidate Aston University Computer Vision

Selected Publications

Here are some of my key contributions to the field of self-supervised monocular depth estimation.

  • BaseBoostDepth: Self-Supervised Monocular Depth Estimation via Effective Baseline Boosting
    • Kieran Saunders, George Vogiatzis, Luis J. Manso
    • British Machine Vision Conference (BMVC) 2024
    • Highlight: Introduced a novel training strategy to leverage larger temporal baselines for improved depth accuracy and detail, achieving SOTA results.
    • [ Coming Soon: Paper, Project Page, Code ] (Update links when available)

  • Self-supervised Monocular Depth Estimation: Let’s Talk About The Weather
    • Kieran Saunders, George Vogiatzis, Luis J. Manso
    • IEEE/CVF International Conference on Computer Vision (ICCV) 2023
    • Highlight: Developed Robust-Depth, enhancing model resilience to adverse weather and visual conditions via augmentation-based pseudo-supervision.


  • Dyna-DM: Dynamic Object-aware Self-supervised Monocular Depth Maps
    • Kieran Saunders, George Vogiatzis, Luis J. Manso
    • International Conference on Autonomous Robot Systems and Competitions (ICARSC) 2023
    • Highlight: Focused on improving depth estimation in dynamic scenes by identifying and handling moving objects.