 |
Hand tracking and recognition of bimanual movements
This project aims to recognise the bimanual gestures. Activities where bimanual movements are important include clapping, sign language, opening a wine bottle, typing on keyboard, eating with knife and fork, drumming, guiding a pilot driving an aircraft to the parking, etc. Tracking the hands in these type of movement is crucial. Also, in some applications like sign language there are many different bimanual movements. A system is required to recognise different movements in discrete and continuous manners.
For more information about this project please see the article in the ERCIM News research magazine.
|
|
  
Real-Time gesture
recognition of human hand
We aim to build a system which can recognise a significant number of signs from a sign language for the Deaf
such as ISL (Irish Sign Language). The system should be able to run on a standard PC and use only a single
webcam. The webcam feeds live video images to the PC where they are processed.
The multi-scale approach works by using the idea of "divide-and-conquer". In this approach every single image is blurred at different stages in order to make a hierarchical database of shapes.This technique reduces the time for searching the known hand shapes dramatically. A hidden Markov model uses the hierarchical database to extract the shapes for recognition.
To see more details about this project click HERE
|
 |
A Gaussian Graph Matching Algorithm for Spatio-temporal Gesture Recognition
In this algorithm we aim to match a given unknown gesture with one of the gestures in the vocabulary. We form a vocabulary of hand gestures and train the system by a set of predefined gestures. We model the gestures by gaussian graphs and develope a Graph Matching algorithm to find the best match of the bipartite graphs formed by the trained gaussian graphs and the graph of the unknown gestures.
Please visit HERE for more detail.
|
|
A Human-Computer Natural Interface
We have developed a Human Computer natural interface for interacting with computers by hand gestures. In order to interact with computers with no physical connection a gesture recognition system is required. In this project we control a mouse pointer by hand gestures and communicate with a PC. This system uses statistical pattern recognition techniques and state machines in order to enable the user do all the tasks provided in an MS Windows environment such as moving, clicking, selecting, cutting and pasting, etc.
Please find a demo of this project HERE
|
 |
Classification for recognition of hand-written digits
A new statistical approach which is based on a hierarchical database of hand written digits is used to recognise the digits.
|
