The Computer Vision Group, as part of the Centre for Computational Engineering Sciences, investigates and develops vision-based solutions across a broad range of applications. The Group is committed to leveraging the latest innovations in the fields of computer vision and machine learning, coupled with experience of delivering effective solutions for both academic and industrial partnerships.
Vision
Computer vision plays a vital role in the day-to-day lives of people across the world. From analysing environmental data captured by remote sensing satellites to improving the ways in which we interact with computer-based systems, computer vision enables novel and robust solutions for a wide range of challenges.
Using cutting-edge machine learning coupled with comprehensive image analysis techniques, we provide solutions across the transport, security and engineering sectors. Our commitment extends beyond technological innovation to encompass ethical and sustainable practices, ensuring our work contributes positively to society and the environment.
We thrive on collaboration, engaging with partners across academia and industry, to drive forward the boundaries of what is possible, making our solutions more accessible and inclusive. By embracing adaptability and focusing on scalable approaches, we are poised to respond to both current and future challenges.
Research topics
We have a long history in computer vision, building on a successful MSc course in Computer and Machine Vision that has run for over 30 years.
Our approach to developing solutions is based on a strong commitment to understanding the requirements of the problem and generating an iterative approach to delivery. We have extensive experience with the integration of systems from different partners, and we rely on robust engineering practices to deliver an effective end product.
We understand that solutions need to be aligned with the primary project goals, the constraints of the operating environment and commercial considerations. To this end, we are committed to providing working solutions on platforms at all levels, from the edge to the cloud. We recognise that providing a pathway to scalability is an essential factor in developing successful solutions, preparing them for future exploitation.
Our expertise and experience include:
Automotive applications
- Analysis of in-vehicle driver behaviour using gaze tracking to monitor attention
- Optimisation of autonomous vehicle awareness systems using simultaneous localisation and mapping (SLAM) methods
- In-vehicle driver support, highlighting risk awareness
- Semantic segmentation for planetary rover navigation
Security and surveillance
- Threat detection via human pose estimation
- Multi-modal sensor fusion and automated threat detection (MoD Grand Challenge winner)
- Vision-based approaches to traffic management, including automated vehicle detection and risk assessment
- Bird strike detection
Human computer interaction (HCI)
- Gesture recognition
- Human pose estimation
- Data visualisation
Advanced computer vision
- Automated training, validation and testing of ML-based vision pipelines
- Active learning in vision applications
- Synthetic image generation and improvement
Industrial partners
The list of our industrial partners continues to grow, and includes Jaguar Land Rover, Airbus, International Coatings, Thales and Blue Bear Systems. We continue to forge partnerships with companies that have an interest in the processing and analysis of visual data and its application to real-world problems.
How to collaborate
MSc student projects provide a low risk, potentially high reward path for exploring ideas and solutions without full commitment of a company's resources. As part of the MSc programme, our students generally embark on projects that are aligned with the research activities of the Centre's academic staff, which can include actively sponsored projects from industry. However, we also welcome suggestions for less formal partnerships, as this gives the student an opportunity to work on industrial projects and gives them experience that is important for their future careers.
For longer-term investigations, funding a PhD student can provide a stable and cost effective approach. The extended period of research brings more depth and insight to the subject, with opportunities for joint supervision to ensure that the goals of the industrial partner are maintained.
We are enthusiastic about forming partnerships that might lead to future funding via UK and EU sourced grants. Cranfield University has a team dedicated to supporting applications to the various funding bodies, such as Innovate UK and UKRI, and has an excellent track record in attracting successful funding.
