Deep Learning Gives DeepHand System a Hand

Deep Learning Gives DeepHand System a Hand
Jul-24-16
The DeepHand system takes advantage of deep learning to track hand motions in real time and display a 3D representation.

Developed by a team from Purdue University, the system uses commercial 3D sensors and a deep learning "convolutional neural network” to detect the position of different points on the hand. That data is then run through an algorithm that chooses the best matching image of a hand from a library of more than 2.5 million poses. Also, by identifying what the team calls the “spatial nearest neighbors,” the DeepHand system can predict what the next hand position in order to display the images more quickly.



More Info about this Invention:

[DATAVERSITY.NET]
[ENGINEERING.PURDUE.EDU]
Next Invention »
Share on      

Add your Comment:

[LOGIN FIRST] if you're already a member.

fields are required.



Note: Your name will appear at the bottom of your comment.