[Kde-accessibility] Lip Reader Demo

Yash Shah blazonware at gmail.com
Tue Mar 20 22:47:33 UTC 2012


Hey everybody,

(Ignore my previous email)

I just wrote demo code of how to track out lip movements to know whether a
person is speaking or not.

Here's some code: https://github.com/yashshah/LipReader (rather hackish
right now).

And here's a video: http://www.youtube.com/watch?v=-fpkZBYBsuc

It detects the change in pixel density in the mouth region between two
consecutive frames and if this is above a certain threshold, it assumes
that the person is speaking. This is not super-accurate, but seems 'good
enough' for a start. This could be refined further in the project later.

In the video, note how it automatically 'locks on' to the new face if a new
person comes in front of the camera and starts reading his lips.

The actual face detection takes place only once every two seconds to save
CPU time. The rest of the time, it only needs to 'track' the face object
using the CamShift algorithm, which is super fast and lightweight and works
by tracking difference between consecutive frames.

This is with respect to the project about face detection for Simon.

Regards,
Yash Shah
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kde-accessibility/attachments/20120321/201e9ef8/attachment.html>


More information about the kde-accessibility mailing list