Knowledge Transfer Partnerships (KTPs) are a unique UK-wide activity that help businesses to improve their competitiveness and productivity by making better use of the knowledge, technology and skills within universities, colleges and research organisations.
Further information is available at: https://iuk-ktp.org.uk/
THE PROJECT
The University of Essex in partnership with Railscape Limited offers an exciting opportunity to a graduate with the relevant skills and knowledge to develop an intelligent 3D visualisation platform that provides personalised, accessibility-aware 3D walkthroughs of rail stations and other major infrastructure sites using advanced spatial computing and AI.
DUTIES OF THE POST
The duties of the post will include:
- Using 3D spatial cameras for modelling railway station infrastructures / facilities to visualise passenger routes and identify objects, assets and areas posing accessibility issues for customers.
- Develop digital 3D mapping tools for capturing passenger routes based on captured video, dense point cloud, spatial and localisation data/processes for generating interactive 3D video guides / walkthrough of railway stations
- Develop and optimise realistic 3D Gaussian Splat models of railway stations / assets from captured 2D photos and videos, ensuring visual quality, spatial accuracy, performance across diverse computationally constraint hardware platforms and interfaces
- Apply state of the art machine learning models for object detection, scene segmentation with the integration of sensor data on operational / usage patterns to identify accessibility issues and risks to customers
- Integrate conversational AI into assistive digital applications for generating personalised, location aware interactive immersive visual walk throughs and accessible route suggestions based on customers’ needs and circumstances.
- Building 3D visualisations for augmented / virtual reality-based interfaces
- Working both independently and as part of a team as both a project manager and a lead researcher
- Balancing innovation with practical implementation
- Working collaboratively with operational staff, senior management, software teams and external rail stakeholders
- Embedding technology and upskilling company staff
- Participating in academic and/or industrial conferences and other events to disseminate research outcomes
- Preparation of academic papers and patent applications
KEY REQUIREMENTS
- A minimum of a Master’s degree in AI, Computer Science Robotics, Mechatronics or Electronic Engineering (Essential)
- Practical and theoretical knowledge of integrating 3D spatial camera features (Gaussian Splat, LiDAR / rendered point cloud data, SLAM, geospatial data) for generating interactive 3D visual scenes
- Practical and theoretical knowledge of advanced deep learning approaches applied to computer vision and multimodal data processing
- Knowledge and experience of developing generative AI applications
- Experience of applying data fusion and machine learning techniques to data rich problems
- A high level of ability in the use of Python
- Additional experience with C++ (desirable)
- Additional experience with CUDA / GPU optimisation (desirable)
- Experience of mobile / web application design and development
- Excellent communication skills and the ability to convey ideas and concepts to a variety of key stakeholders
- The ability to act as both a project manager and a lead researcher
- The ability to work both independently and as part of a team
LOCATION
Railscape Limited, 15 Totman Crescent, Rayleigh, Essex, SS6 7UY
Please use the 'Apply' button to read further information about this role including the full job description & person specification which outlines the full duties, skills, qualifications & experience needed for this role. You will also find details of how to make your application here.
Our website http://www.essex.ac.uk contains more information about the University of Essex. If you have a disability & would like information in a different format, please email .
:quality(80))
:quality(80)){kind=icon}
:quality(80))