Face detection tracking and recognition for a visual prosthesis
The present invention is a system for detecting, tracking and recognizing human faces in a visual prosthesis. In a visual prosthesis, the input camera is always higher resolution than the electrode array providing percepts to the subject. It is advantageous to detect, track and recognize human faces. This information can be provided to the subject by highlighting the face in the visual scene, providing auditor or vibratory notice that a human face is in the visual scene, looking up the face in a database to identify and annunciate the name of the person in the visual scene, or otherwise communicate the identity of the person, like providing a custom vibratory pattern corresponding to known individuals (like custom ring tones associated with caller ID) or automatically zooming in on a face to aid the subject in identifying the face.
1. A method of aiding a visual prosthesis user by performing, in a video processing unit, the steps comprising:
analyzing a visual scene from a video input device, the video input device having a field of view;
receiving data from the video input device and generating stimulation data;
detecting a face in the field of view, with the video processing unit yielding a detected face;
providing a vibration cue to the visual prosthesis user when a face is detected;
providing an indication of the location of the detected face within the field of view; and
stimulating visual neurons of the visual prosthesis user with the stimulation data from the video processing circuit.
2. The method according to claim 1 , further comprising looking up the face in a look up table to retrieve a vibration pattern associated with the detected face and cueing with the vibration pattern associated with the detected face.
3. The method according to claim 1 , further comprising detecting a characteristic of the detected face selected from a list consisting of gender, age, ethnicity, size, distance, direction of movement, and facing direction, and including information about the characteristic in the vibration cue.
4. The method according to claim 1 , wherein the cue further includes an audio cue.
5. The method according to claim 2 , wherein the cue further includes stating a name associated with the detected face.
6. The method according to claim 1 , wherein the cue further includes highlighting the detected face in the stimulation data.
7. The method according to claim 3 , wherein the cue further includes an indication of if the face is looking toward the visual prosthesis user or turned away.
8. The method according to claim 1 , wherein the cue further includes tactile feedback.
9. The method according to claim 3 , wherein the characteristic is gender.
10. The method according to claim 3 , wherein the characteristic is size.
11. The method according to claim 3 , wherein the characteristic is distance.
12. The method according to claim 3 , wherein the characteristic is head movement.
13. The method according to claim 3 , wherein the characteristic is motion.
14. The method according to claim 1 , further comprising providing controls on the video processing unit to control the step of detecting a face.