Sunday

New Shoes?


Public technology, such as table top interactive screens offer a great way to collaborate, work in groups and access information in an intuitive fashion.

Personalising content to these devices is difficult, either requiring users to login via screen input or the use of proximity based ID which the users carry on their person. Other methods such as face recognition have been used in the past - however normally only to identity gender rather than actually identifying specific users.

Kinect

A novel approach to this problem has been developed by Christian Holz at the Human-Computer Interaction group using Microsoft Kinect cameras.

The cameras mounted to the base of the table scan users shoes, comparing them to a database of profiles - matching users to pre registered shoes.

Users interactions with the table are also associated to the relevant user (when groups are interacting with the table) by matching hand orientation with users shoes.

Two left feet

Obviously shoe size and type offer only a certain amount of uniqueness, especially in an environment such as a university where Converse are the footwear of choice. Challenges such as these could be over come through menu options allowing users to choose their profile from a list of possible candidates.

Footfall

Another possible application for the technology could be to better measure footfall in department stores. Todays technology gives indication of the number of people entering the store, this technology could distinguish shoe size, giving a better indication of adults VS children and therefore a more realistic indication of retail opportunities.


links : http://www.hpi.uni-potsdam.de/baudisch/projects/bootstrapper.htm

No comments: