The role
This role will work on the specific project "PriorPool: Intelligent Video Restoration And Enhancement Via A Large Prior Database" funded by EPSRC. This project assembles interdisciplinary experts in image processing and computer vision (University of Bristol), media content delivery (BBC), filming equipment and services (Esprit Film & Television), and post-production technologies (Lux Aeterna). We pioneer new AI-enabled techniques to enhance the quality of videos captured in challenging environments.
PriorPool aims to develop an advanced framework to improve video quality, specifically targeting footage from natural settings that often suffer from environmental distortions. Video distortions can dramatically alter data characteristics, leading to significant discrepancies between the representations of clean and distorted content. Your role will involve diving deep into the world of video restoration and enhancement, focusing on leveraging well-defined prior knowledge obtained from diverse and representative datasets. The new methods not only recognise the video distortions but also adapt effectively to ensure the highest fidelity in video enhancement.
What will you be doing?
Research Responsibilities
1. You'll hypothesise and model complex video restoration techniques that require a precise understanding of the content and its context.
2. You'll tackle the challenge of learning from limited high-quality production videos by developing a sophisticated query system. This system will analyse distorted videos and fetch corresponding priors from high-quality sources to aid in their restoration.
3. You'll engage in cutting-edge research, employing unsupervised learning methods and unsupervised domain adaptation to bridge the gap between distorted and pristine video content.
4. You'll publish findings in high-impact journals and international conferences.
Administration Responsibilities
The post holder will be expected to contribute to the organisation of activities, including those associated with the programmes mentioned above.
You should apply if
You have:
1. A strong research track record in image processing; and/or computer vision; and/or in deep learning with application to video enhancement problems and an ability to innovate.
2. Proven research experience in image processing and in deep learning and/or related areas.
3. Strong project management skills.
4. A commitment to share knowledge in an enthusiastic and effective manner.
5. Proven ability to publish papers in top venues relevant to the area of research.
6. Experience in delivering successful research projects.
7. A proven track record of collaborating with both academic and industrial partners.
Additional information
For informal queries please contact:
Dr Pui Anantrasirichai - n.anantrasirichai@bristol.ac.uk
Contract type: Open ended with fixed funding for 27 months
Work pattern: Full time
Grade: I / J
Salary: Grade I £37,999 - £43,878 per annum; Grade J £42,632 - £47,874 per annum
School/Unit: School of Computer Science
This advert will close at 23:59 UK time on 17th March 2025.
Interviews are expected to take place week commencing 1st April 2025. #J-18808-Ljbffr