VidGrid: Distributed video analysis with grid technologies
VidGrid was a 21-month project (January 2004 - August 2005) funded by the ESRC, under the e-Social Science pilot projects scheme. This funding constituted part of the ESRC's involvement in the UK e-Science programme, coordinated by the National Centre for e-Social Science (NCeSS).
Please note: VidGrid has now completed, but the software we have developed will now be developed into a full-scale publicly available version through the e-Social Science node 'MiMeG'.
e-Science tools and infrastructures offer the potential to support the growing interest in the use of audio-visual recordings as an analytic resource for social science. Technologies for recording, replaying and remotely sharing digital video materials have increased in sophistication, availability and simplicity. As a result, a growing number of distributed multidisciplinary teams work with video data. The project is investigating the potential for Grid technologies to support and enhance the collaborative analysis of video data. For example, the availability of high-performance resources could support real-time video sharing and manipulation; and the use of metadata can enable innovative forms of corpus visualisation and data markup (analytic annotations and the like).
We have running prototype video data delivery and annotation services to support distributed video analysts.
In parallel, the project is generating requirements for video analysts through focused studies and workshops. Through such work we are demonstrating the capacity of infrastructures to genuinely support qualitative social science, whilst simultaneously scoping the key constraints of such a process.
Deliverables
Best, K., and the VidGrid Consortium, State of the Art Software Supporting Video Analysis. [PDF]
Publications
Fraser, M., Biegel, G., Hindmarsh, J., Heath, C., Best, K., Reeves, S. and Greenhalgh, C., Object-Focused Interaction in e-Social Science, in Proceedings of Workshop on Social Aspects of Scientific Collaboration at the First International Conference on e-Social Science, Manchester, UK, 22nd-24th June 2005 [PDF]
Fraser, M., Biegel, G., Best, K., Hindmarsh, J., Heath, C., Greenhalgh, C. and Reeves, S., Distributing Data Sessions: Supporting remote collaboration with video data, in Proceedings of the First International Conference on e-Social Science, Manchester, UK, 22nd-24th June 2005 [PDF]
Fraser, M. and the VidGrid Consortium, The Same Old Remote Misunderstandings: Object-Focused Interaction in e-Social Science, in Proceedings of Ubicomp 2004 workshop on Giving Help at a Distance, Nottingham, UK, 7th-10th September 2004 [PDF]
Presentations
Agenda Setting Workshop on Collaboration and Co-laboratories, June 2005
Workshop on Video, Social Science Research and Technical Innovations, March 23rd 2005 [PPT]
National Centre for e-Social Science All Hands Meeting, Manchester, July 5th 2004 [PPT]
UK e-Science All Hands meeting, September 1st 2004 [PPT]
Object-Focused Interaction in e-Social Science, September 7th [PPT]
Many other presentations, please contact Mike for slides
Workshops
Workshop on "Bringing data together: interactions, logs, and requirements", Nottingham, UK, in collaboration with the Equator IRC, July 5th 2004
Workshop on "Video, Social Science Research and Technical Innovations", London, UK, March 23rd 2005
Various other workshops, contact Mike for details
Involvement during the project
Dr. Mike Fraser [contact]
Dr. Greg Biegel
Dr. Jon Hindmarsh
Katie Best
Prof. Christian Heath
Stuart Reeves
Prof. Chris Greenhalgh
Prof. Steve Benford
Partners
Mobile and Wearable Computing Group, Department of Computer Science, University of Bristol
Work, Interaction and Technology group, Management Centre, King's College London.
Mixed Reality Laboratory, University of Nottingham
What is VidGrid?: some alternative views
The "we should have picked a less popular project title" section.
VidGrid is program useful for studying motion visually
VidGrid is also a rack-mountable server