Kinsight is an experimental system that uses networked Kinect cameras to locate missing objects within a household
Have you ever wished that your computer could answer questions like “What did he mean when he said that?” or “Should I go to that party?” Well, while that may never happen, it could conceivably provide you with an answer if you asked it, “Where did I leave my wallet?” – if it were running the Kinsight system.
Kinsight was developed by Shahriar Nirjon and John Stankovic, who are computer scientists at the University of Virginia in Charlottesville.
The experimental system incorporates networked Microsoft Kinect depth-sensing cameras placed within every room of a house. Those cameras track people as they move through the rooms, and note the size and shape of whatever items they cause to change locations. That information is cross-referenced with a database of the size and shape of objects that belong in the house, allowing the system to determine what things have been moved to what places.
If a user of the system were not able to find something, they could essentially “search” Kinsight using that item’s name, to get its last known location. Apparently, it is able to track fist-sized objects to within a range of about 13 centimeters (5 inches).
In order to be searchable, however, items must first be tagged by the user – Nirjon is currently working on an app to make that process easier.
Source: University of Virginia
Kinsight was developed by Shahriar Nirjon and John Stankovic, who are computer scientists at the University of Virginia in Charlottesville.
The experimental system incorporates networked Microsoft Kinect depth-sensing cameras placed within every room of a house. Those cameras track people as they move through the rooms, and note the size and shape of whatever items they cause to change locations. That information is cross-referenced with a database of the size and shape of objects that belong in the house, allowing the system to determine what things have been moved to what places.
If a user of the system were not able to find something, they could essentially “search” Kinsight using that item’s name, to get its last known location. Apparently, it is able to track fist-sized objects to within a range of about 13 centimeters (5 inches).
In order to be searchable, however, items must first be tagged by the user – Nirjon is currently working on an app to make that process easier.
Source: University of Virginia
0 comments:
Post a Comment