Adobecomputer visionimage trackingPedestrian Tracking

Video Tracking and Manipulation: Adobe and University of Washington

By November 30, 2008 No Comments

Video tracking has huge potential for communicating issues of place and space in the urban environment as well as for data tagging and data collection for use in a GIS. Last weeks post on tracking pedestrians highlighted what could be archived with a simple webcam, if you add in computer vision and on the fly path analysis then you take the concept a step further.

Three dimensional video tracking has been around for a while but it is generally restricted to high end packages. If you have ever wanted to tag data onto video objects, analyse traffic flows and pedestrian paths or augment scenes then the video below brings it a step closer to consumer level:

Interactive Video Object Manipulation from Dan Goldman on Vimeo.

For more information take a look at the following publications:

Goldman, D. B, Gonterman, C., Curless, B., Salesin, D., and Seitz, S. M. 2008. Video object annotation, navigation, and composition. In UIST ’08: Proceedings of the 21st annual ACM symposuim on User Interface Software and Technology, 3–12.

Goldman, Daniel R. 2007. A framework for video annotation, visualization, and interaction, PhD Thesis, University of Washington.Goldman, D. B., Curless, B., Seitz, S. M., and Salesin, D. H. 2006. Schematic storyboarding for video visualization and editing. ACM Transactions on Graphics (Proc. SIGGRAPH), 25(3), 862–871.


Andy is Professor of Digital Urban Systems at the Centre for Advanced Spatial Analysis at University College London.