| Autor: |
Pardhu, Thottempudi, Deevi, Nagesh, Rao, N. Srinivasa |
| Předmět: |
|
| Zdroj: |
AIP Conference Proceedings; 2024, Vol. 3028 Issue 1, p1-7, 7p |
| Abstrakt: |
Hand signal acknowledgment in view of man-machine association point is being developed quickly lately. Due to the effect of lighting and complex establishment, most visual hand signal acknowledgment structures work simply under restricted climate. A versatile skin shading model in light of face location is used to distinguish skin shading locales like hands. To arrange the powerful hand signals, we developed a basic and quick movement history picture-based technique. Hand gestures is one of the most reliable medium languages between the people who are deaf and mute and that use the visual-manual view to convey meaning. In this project, we take input of the hand gestures from a camera and do the video processing i.e. frame per second, using python and decode the hand gestures using the previous input which was given to the training module and the given distinct images goes through a number of training steps in improving the output enhancement for the real time output data i.e. this was carried out with the assistance of the Media Pipe, which is based on the customizable ML solutions for live and streaming media and in this project this helps with the concept of Landmark points/Check points that indicate all the joint points in hand ligament and with each formation of the hand sign according to the check points be labeled to it's respective sign, which are located with the numerous number of dataset i.ie which collects all the subsequent key point coordinates and it'll be trained in using keras and tensor flow for the training of the each and every hand gesture that's given as an input in the real time. There are different hand sign gestures that are used in different countries according to the need and the understanding that they can lean on and out of all the most commonly used language is IG i.e. International gestures and also ASL i.e. American Sign Language is used, since it was generally utilized in the English i.e. International language and the problem which we wanted to solve in our project is that the program can adapt to the related user that can be used to the user application and its application can be implemented in various work spaces i.e. in start to finish video calls, expanded reality, impeded, games. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
