On 9/2/2010 9:59 AM, Benjamin Scott wrote: > I agree. First, you need to be able to detect people. > Second, Detect Items in their hands. Third, Where the > hand move to. Fourth, Flag all "impacts" that don't > contact hands: Such as handshaking.
It should be pointed out that each one of these things is a likely candidate for a semester-long project in and of themselves. Analyzing 2D camera data in real-time is a rather difficult undertaking. > --- On Wed, 9/1/10, Santosh Vernekar<[email protected]> wrote: >>> First the algorithm has to be understood, programming >> comes later. >> You might want to first look as basic video analysis >> algorithms. (eg: >> identifying a disturbance by analyzing video pattern change >> etc.) >> Some of these basic features are already present in >> surveillance cams. >> >> However this definitely is off topic here, for c-prog ML ! What programming language do you expect the OP to develop this project in? A quick couple searches turned up: http://sourceforge.net/projects/opencvlibrary/ OpenCV looks like an active SF project in wide use. Has C/C++ bindings. Looks like a good library to start from. Better than trying to roll your own object detection algorithms from scratch. -- Thomas Hruska CubicleSoft President Barebones CMS is a high-performance, open source content management system for web developers operating in a team environment. An open source CubicleSoft initiative. Your choice of a MIT or LGPL license. http://barebonescms.com/
