UCL Department / Division
Civil, Environmental & Geomatic Engineering
Location of position
London
Duration of Studentship
4 Years
Stipend
The PhD is a 4-year fully funded position (tuition and a stipend at the Research Council rate) plus a CASE supplement)
Vacancy Information
Applicants are sought for an innovative PhD studentship with University College London (UCL) and Ordnance Survey. The project aims to enhance geospatial capabilities to automatically process and understand multi-domain point cloud data. This challenging research will investigate the latest approaches for 3D data machine learning across multiple domains (sensor domain, data domain).
Studentship Description
Point clouds are one of the most significant data representations for sensed 3D data and underpin applications in mapping, digital engineering, digital twins, urban analytics, robotics, autonomous driving, and mixed reality. The data acquisition methods and devices, such as photogrammetry and LIDAR, have reached a mature state. Therefore, point cloud data is widely available or can be acquired at moderate cost.
This is contrasted by a relative Immaturity of algorithms that can automatically understand and classify 3D sensed data with high reliability or accuracy. Over the past few years, deep learning emerged as a key driver for progress in 3D point cloud understanding. Next-generation AI based on Deep Learning promises to be a game changer in automated 3D point cloud understanding.
UCL Geomatics in collaboration with Ordnance Survey (OS) will investigate AI technology to automatically understand 3D point cloud data aiming for label-efficient methods. We propose to investigate:
- Label transfer from existing GIS data to point cloud data, from existing data domains to new data domains and from existing sensor data to new sensor data
- Self-supervised methods (for pre-training) such as contrastive learning
- Next-generation deep learning architectures such as self-attention networks, sparse conv and point cloud transformers.
The project is a collaboration between Ordnance Survey, Britain’s National Mapping Agency and University College London, in particular UCL Geomatics. OS has expertise in geospatial data capture and processing, including photogrammetry and image processing. Their research and innovation teams have developed machine learning techniques to extract information from aerial and satellite imagery. As 3D data becomes more widely available it will allow OS to extract more information from its data. UCL Geomatics has long established research expertise in photogrammetry and LIDAR data processing; 3D point cloud generation and processing; object recognition from point clouds and imagery and geospatial data analytics. The academic supervisor will be Dr Jan Boehm.
Person Specification
A successful applicant is expected to have completed an excellent University degree in Geomatic Engineering, Computer Science, Data Science or a related field. The applicant should have a strong background in at least one of the following areas: geomatics, photogrammetry, LiDAR, computer vision, machine learning, data science. Candidates should have excellent communication and presentation skills and will ideally be highly motivated to apply academic research to real-world problems.
Eligibility – Home/Overseas
To be eligible for fees at the UK rate, you must normally be a national of the UK (or in specified cases the family member of a UK national), be ordinarily resident in the UK on the first day of the first academic year of your programme and have been ordinarily resident within the UK, the Republic of Ireland, the specified British overseas territories or the Channel Islands/Isle of Man (the “Islands”) for the three year period before the first day of the first academic year of your programme.
Applicants should send a covering letter and CV to Jan Boehm ([Email Address Removed])
The successful applicant will then have to apply online to UCL by submitting the PhD application form available via Civil, Environmental and Geomatic Engineering MPhil/PhD | UCL Graduate degrees - UCL – University College London. Please name Jan Boehm as the proposed supervisor.
Contact name
Dr Jan Boehm
Contact details
[Email Address Removed]
Closing Date
30 June 2022
Interview date
TBC
Studentship Start Date
1 October 2022