Latest Bristol Vision Institute News
29 October 2012
3D, Robot Vison and video compression.
Bristol-BBC Immersive Technology Laboratory- Beyond 3D!
Funded by EPSRC, this new research collaboration builds on the strengths of Bristol Vision Institute and BBC R&D with associated partners including BBC NHU and Aardman. The work spans engineering (Bull, Thomas (BBC)) and psychology (Gilchrist) with application to the creative arts (Flaxton). The research will provide key underpinnings for future consumer and professional technology and services, in particular related to the rate-quality trade offs for new high frame rate and high dynamic range display formats. The new facility supports a subjective testing laboratory which (based on the work of Troscianko and Hinde) will, for the first time enable accurate assessment of the presence created by a particular format and display environment. For further details contact David Bull [firstname.lastname@example.org].
Biped Robot Vision
A remarkable aspect of animals as ‘biomechatronic’ systems is the performance of their control system given its relatively simple sensors. Although the technology exists to produce image sensors that can outperform the eyes of humans and cheetahs, there is no artificial approximation of the visual perception capacity of animals. A £550k EPSRC research grant has been awarded to BVI (Burn, Bull, Mayol-Cuevas and Gilchrist) to investigate the use of artificial visual perception for control of locomotion in legged robots. There is much known about the way humans use vision to control locomotion. This has however, yet to be translated to the engineering disciplines of machine vision and robotics. For further details contact J. Burn [email@example.com].
EPSRC funding for BVI in video compression
The predicted growth in demand for bandwidth, especially for mobile services will be driven by video. David Bull (VI-Lab), Dimitris Agrafiotis (VI-Lab) and Roland Baddeley (Experimental Psychology) have won a new £600k research grant to investigate perceptual redundancy in, and new representations for, digital video content. In collaboration with BBC and HHI-Fraunhofer Berlin, the team will investigate video compression schemes where an analysis/synthesis framework replaces the conventional energy minimisation approach. Early results by Zhang and Bull, modelling scene content using computer graphic techniques is already producing world-leading results and has the potential to create a new content-driven framework for video compression. For further details contact David Bull [firstname.lastname@example.org].