Entraides et échanges autour de la technologie Scol - Informations and exchanges on the Scol technology
Vous pouvez changer la langue de l'interface une fois inscrit - You can change the language once registered
You are not logged in.
Pages: 1
Hello Arkeon,
A very interesting resource for the development of educational aplicarivos is the ability to interact with the virtual world with your hands, but Leap Motion hardware only works on the desktop environment (Windows / Linux / Mac).
I wonder about the feasibility of implementing the recognition of hands using computer vision, so that the resource can be exported to Android. There are some jobs available on the internet as referênca: https://ilab.cs.ucsb.edu/projects/taehe … ndyAR.html
The recognition feature of the hands need not have recognition of hand pieces, such as the phalanges of the fingers, simply a more general recognition, allowing just touch the virtual objects using the physical hands.
Thanks,
Alexandre - Brazil
Offline
I'm not sure how robust this is.
The skin detection and background removal technique followed by hand contours detection work a lot better when using depth camera.
To make computer learn hands and fingers like for the leap motion you needs neural networks again with a lot of data.
I know that similar devices (Depth camera) for smartphone and tablets exist, but I don't know if this is actually available.
Offline
Got it. The smatphones does not have sufficient processing for background extraction operations and other operations required for detecting a hand. We have to wait for the evolution of processing power such equipment.
Thanks
Offline
Offline
Pages: 1