Multimedia computing

Our focus is to develop multimedia technologies that enhance human experience and performance.

Multimedia data, such as text, audio, images, animations and video, have become an integral part of everyday life.

We conduct research on multimedia content understanding, creation, analysis, and processing. We focus on developing technologies and systems that enhance human experience and performance, and that tackle the challenges faced by online content service providers.

Multimedia computing also benefits a wide range of applications, including health, telecommunication, security, and education.

The group has a track record of delivering impact through research. Our research combines mathematical models, artificial intelligence, machine learning techniques, and human factors in algorithm development and evaluation.

The expertise spans a broad range of topics, including image processing, computer vision, augmented/virtual reality, computational music, speech analysis, and visualisation. We have a track record of collaboration within Cardiff University and externally.

Aims

Develop algorithms for analysing, processing and assessing multimedia data.
Develop systems that enhance human experience and performance.
Develop technologies that solve intellectual, industrial, and societal challenges.

Research

Research themes

Our research fits broadly within the following themes:

Multimedia experience

Image and video quality assessment.
Eye tracking and saliency modelling.
Perceptual quality of images/3D data.

Multimedia content generation and analysis

Content generation for images and videos.
3D data (point clouds, meshes) content generation.
Synthesis of realistic humans/faces and caricatures.
Computational music.
Speech analysis.

Engaging users with multimedia

Emotional and social signals.
Sentiment analysis for multimedia data.
VR/AR techniques.
Human factors in visualisation.

Projects

Funding

Project name: Integration of video realistic avatar capability into an artificial intelligence driven healthcare information platform
Funded by: KTP
Principal investigator: Professor David Marshall

Project name: Seeing the future

Funded by: EPSRC

Principal investigator: Professor David Marshall

Project name: Perceptually Salient Video Quality Awareness via Scene-Level Quality Assessment

Funded by: Royal Society

Principal investigator: Dr Hantao Liu