Proj-2013-2014-Sign2Speech-English
Objective
The goal of our project is to allow speechless people to communicate with a computer thanks to sign langage. Our programm should be able to understand sign langage in order to obey to orders given. Then orders should be displayed under literal form and transformed into vocal signal thanks to a speech synthesis technology.
The programm should also be able to learn new hand movements to increase its database to recognize more ideas.
Software Requirements Specification
The Team
- Tutor : Didier Donsez
- Members: Arthur CLERC-GHERARDI, Patrick PEREA
- Department : RICM 4, Polytech Grenoble
State of the Art
Recognition of sign langage alphabet
Recognition of particular hand movements which mean ideas (ZCam camera)
Recognition of particular hand movements which mean ideas, and the traduction in spanish (OpenCV)
Recognition of particular hand movements which mean ideas (Kinect)
Tools
We will use two different technologies :
The Leap motion is a device that allow you to control your computer with your hands. But there is no physical contact and the communication with the computer is based on hand movements. You put the Leap motion under your hands, next to the keyboard.
Compared to the Kinect, it is much smaller. The device has a size of 8 x 2,9 x 1,1 cm and a frame repetition rate of 200 Hz (the Kinect has a 30 Hz frame repetition rate). The Leap motion is composed of two webcams of 1,3 MP which film in stereoscopy and three infra-red light LED. It can detect the position of all ten fingers.
The official website has a section for developer. You can download the SDK 1.0 (almost 47 Mo) which contains API for following langages: C++, C#, Java, Python, Objective C and JavaScript. The SDK also contains examples for libraries and functions of the SDK.
- The Creative camera of Intel® Perceptual Computing SDK
This camera is also a remote controller for the computer. You put it in front of people.
The Creative is provided of depth recognition, which enable developer to do the difference between shots. This camera films at around 30 fps in 720p.
Intel provides also a SDK for developers : [1]. Some of the libraries will help us for hand and fingers tracking and facial recognition.