| Title |
Rankos gestų atpažinimo, naudojant mašininio mokymosi algoritmus, tyrimas / |
| Translation of Title |
The research of hand gesture recognition using machine learning algorithms. |
| Authors |
Vaitkevičius, Aurelijus |
| Full Text |
|
| Pages |
74 |
| Keywords [eng] |
gesture recognition ; machine learning algorithms, leap motion ; control using gestures |
| Abstract [eng] |
In this paper I analyzed gesture recognition using machine learning algorithms and Leap Motion device. Leap Motion generates a virtual 3D hand model by recognizing and tracking user hands. From this model Leap Motion application programing interface generates a lot of useful date like hand or finger location in 3D space, its vectors and etc. In this paper I presented a system that is capable of learning gestures by passing the data from the Leap Motion device to it. System uses template matching method and “Hidden Markov classification machine learning” (HMC) algorithm. The main use of HMC is two recognize similar data lists. This algorithms always returns a value, so few filtering methods was added to the system. Those methods will not return any value if the gesture is not similar. Experiments was conducted that show has fast the system algorithms can learn to recognize a gesture and how difficult it is to teach the system. All of the results is analyzed in this paper. |
| Dissertation Institution |
Kauno technologijos universitetas. |
| Type |
Master thesis |
| Language |
Lithuanian |
| Publication date |
2016 |
