PhD Candidate in Computer Science, University College London
My research focuses on computer vision and surgical navigation, particularly depth and pose estimation, semantic understanding, and deformable 3D reconstruction for endoscopic navigation. I am supervised by Dr. Francisco Vasconcelos and Dr. Sophia Bano.
I am open to internship and full-time opportunities in computer vision, 3D reconstruction, SLAM, and image-guided surgical navigation.
Computer VisionSurgical Navigation3D Reconstruction
Paper under review
Colonoscopy video understanding and reconstruction through adaptive keyframe selection and structure-aware Gaussian mapping, retaining informative observations while explicitly modelling fold boundaries and temporal deformation.
MICCAI EndoVis Challenge, 2024
The first colonoscopy dataset with detailed pixel-level annotations of colon fold edges and three surgical tool classes, together with benchmarks for segmentation, classification, and annotation-efficient learning.
International Journal of Computer Assisted Radiology and Surgery, 2026
Introduces strong structural and illumination cues for geometry learning in the low-texture colon: intrinsic shading guides depth, while explicit fold-edge features improve self-supervised camera pose estimation.
MICCAI, 2026
Jointly estimates camera motion and deformable anatomy online by anchoring 3D Gaussian Splatting to dense 2D tracks and using motion-aware pose optimization to reduce drift.
ECCV Workshop, 2024
Reveals where standard image-matching benchmarks diverge from downstream structure-from-motion performance, including under domain shift and edge-based preprocessing.
* Equal contribution