I spent a couple of hours yesterday playing with GazeTracker, HeadMouse, and some open-source communication systems. I have two working web-cams.
Most of these systems appear to be designed for adults who probably have longer attention spans than children. I'm thinking there will need to be two ways of tracking a child's movement, no matter what is being tracked. I need little algorithm to take care of this and also to filter the signal from the noise.
I'm trying things on myself, mimicking the movements I know that some of the students make. I like HeadMouse because you can choose what part of the face you'd like to use to click, drag, and open things - the mouth, an eyebrow, an eyeblink- and you can change the settings, too.
To make things work better, I changed the settings within the Accessibility control panel. They would need to be adjusted to meet the needs of the child.
Also, I'd like to try this concept using the SmartBoard as the display. I wonder if anyone has done this? I thought it would be a good way to set up an adaptive cognitive assessment. It is difficult to find appropriate tools for assessing cognitive and learning abilities/skills of students with limited mobility and expressive communication.
1:40 PM
Update: More information about this topic is posted on the TechPsych blog.
I'm working with a few students who have multiple special needs who are in need of assistive communication technologies that are not expensive. Although my first idea was to look at eye gaze technology, the possibility of face tracking is also appealing. Take a look at the following video from SeeingMachines, a commercial company:
I have a couple of webcams, and I don't mind a trip to RadioShack. I can program a bit, too, but I have very limited time. I welcome any advice you might have.
posted by Lynn Marentette at 12:21 PM on Sep 12, 2009
3 Comments
Close this window Jump to comment formthe company menyioned offers a free version - perhaps your project will fall under the rules...?
1:22 PM
It might.
I spent a couple of hours yesterday playing with GazeTracker, HeadMouse, and some open-source communication systems. I have two working web-cams.
Most of these systems appear to be designed for adults who probably have longer attention spans than children. I'm thinking there will need to be two ways of tracking a child's movement, no matter what is being tracked. I need little algorithm to take care of this and also to filter the signal from the noise.
1:32 PM
I'm trying things on myself, mimicking the movements I know that some of the students make. I like HeadMouse because you can choose what part of the face you'd like to use to click, drag, and open things - the mouth, an eyebrow, an eyeblink- and you can change the settings, too.
To make things work better, I changed the settings within the Accessibility control panel. They would need to be adjusted to meet the needs of the child.
Also, I'd like to try this concept using the SmartBoard as the display. I wonder if anyone has done this? I thought it would be a good way to set up an adaptive cognitive assessment. It is difficult to find appropriate tools for assessing cognitive and learning abilities/skills of students with limited mobility and expressive communication.
1:40 PM