Feature-Realistic Neural Fusion for Real-Time, Open Set Scene Understanding (2022)
Real-time Mapping of Physical Scene Properties with an Autonomous Robot Experimenter (CoRL oral, 2022)
iSDF: Real-Time Neural Signed Distance Fields for Robot Perception (RSS, 2022)
iLabel: Interactive Neural Scene Labelling (arXiv, 2021)
iMAP: Implicit Mapping and Positioning in Real-Time (ICCV, 2021)
Incremental Abstraction in Distributed Probabilistic SLAM Graphs (ICRA, 2022)
NodeSLAM: Neural object descriptors for multi-view shape reconstruction (3DV, 2020)
Morefusion: Multi-object reasoning for 6d pose estimation from volumetric fusion
Research Experience
Involved in several research projects focusing on the development of real-time systems and mapping physical scene properties among others.
Education
Pursuing a PhD at the Dyson Robotics Lab, Imperial College London, under the supervision of Prof. Andrew Davison.
Background
PhD student at the Dyson Robotics Lab, Imperial College London. Research interests include Computer Vision, SLAM, Neural Scene Representations, Unsupervised/Continual Learning, and Robotics. Aims to develop methods for 3D scene understanding that enable machines to have meaningful interactions with their environment.
Miscellany
Enjoys building practical real-time systems which work in the real world.