Method and device to speed up face recognition

Patent metadata

Field	Value
Publication number	US-9740913-B2
Application number	US-201113992914-A
Country	US
Kind code	B2
Filing date	Dec 8, 2011
Priority date	Dec 10, 2010
Publication date	Aug 22, 2017
Grant date	Aug 22, 2017

Abstract

Method to customize an application associated with a television experience based on the recognition of users located in front of a display and in the field of view of a camera, comprising the following steps: —an initialization step during which each user is enrolled in a database of a computer system and is defined by a profile referenced by a profile ID and comprising the user name, biometric data and additional personal data, —a finding step during which a wide image, acquired by said camera is scanned to isolate at least one user's faces, to define a marking area surrounding it, to memorize the position of said marking areas, —a matching step to extract the biometric data from said marking area, to match them with the biometric data of the profiles stored in the database, and to assign the detected profile ID with the marking area. While subsequent identification is requested by the application, the following steps are executed —acquiring a wide image with the camera, —extracting the areas of said image according to the marking areas, —extracting for a particular marking area the biometric data of a face, —starting the comparison of the extracted biometric data with the biometric data of the profile ID related to this marking area, and in absence of match, continuing with the other biometric data of the database until one profile is found, —transmitting the found profile ID to the application.

First claim

The invention claimed is:

1. A method to customize an application associated with a television experience based on the recognition of users located in front of a display and in the field of view of a camera, the method comprising:
an initializing operation, during which each user is enrolled in a database of a computer system and is defined by a profile referenced by a profile ID, the profile comprising
a user name,
biometric data of said user acquired by said camera through a face recognition process, and
additional personal data of said user;

a finding operation, during which a wide image, acquired by said camera and embracing a wide area to encompass all potential users located in front of said display, is scanned in order to,
isolate at least one user's face appearing on said wide image,
define a marking area surrounding each isolated face,
store the defined marking area as a fast scanning area image, and
store the position of each of said marking areas in a register that identifies said marking areas;

a matching operation during which each marking area is processed using the stored positions in order to
extract biometric data of the face appearing inside a selected one of the marking areas from the fast scanning area image,
match said extracted biometric data with the biometric data of the profiles stored in the database, and assign the detected profile ID with the selected marking area, and
execute, while subsequent identification is requested by the application, the following operations,
acquiring a second wide image with the camera,
extracting from the second wide image an active area according to the selected marking area,
storing the extracted active area as a second fast scanning area image,
extracting the biometric data of a face from the second fast scanning area image,
starting the comparison of the extracted biometric data with the biometric data of the profile ID assigned to the selected marking area, and in absence of a match, continuing with the other biometric data of the database until one profile is found, and
transmitting the found profile ID to the application.

2. The method of claim 1 , wherein in the case of absence of a match during the subsequent identification, updating the found profile ID related to the selected marking area.

3. The method of claim 1 , wherein in the case of absence of a match during the subsequent identification, updating the found profile ID related to the selected marking area in case of a plurality of detections of the same profile ID.

4. The method of claim 1 , further comprising:
detecting the position of a signal emitted by a remote control within said wide image,
determining the distances between the signal's position and at least one marking area position stored in said register,
identifying the profile ID related to the marking area which corresponds to the shortest distance.

5. The method of claim 1 , further comprising:
comparing at least one personal data common to the found profile IDs, and
assigning a degree of priority to at least one of these profile IDs on the basis of at least one criterion.

6. The method of claim 4 , wherein said signal is an Infra-Red light signal.

7. The method of claim 4 , wherein said signal is a beacon signal specifically dedicated to signal the position of the remote control.

8. The method of claim 7 , wherein said beacon signal has a wavelength in the blue range of a spectrum.

9. The method of claim 7 , wherein said beacon signal has a wavelength in the near Ultra-Violet range of a spectrum.

10. The method of claim 7 , wherein said beacon signal is an intermittent signal.

11. A device to provide a profile identifier ID of at least one user to an application associated with a television experience, comprising:
a database; and
at least one processor connected with a camera, the processor configured to execute an initialization operation during which each user is enrolled in the database of the device and is defined by a profile referenced by a profile ID, the profile comprising the following data,
a user name,
biometric data of said user acquired by said camera through a face recognition process, and
additional personal data of said user,
said at least one processor configured to execute,
a finding and matching operation during which a wide image, acquired by said camera and embracing a wide area to encompass all potential users located in front of said display, is scanned in order to,
isolate at least one user's face appearing on said wide image,
define a marking area surrounding each isolated face,
store the defined marking area as a fast scanning area image,
store the position of each of said marking areas in a register that identifies said marking areas,
extract biometric data of the face appearing inside a selected one of said marking areas from the fast scanning area image, and
match said extracted biometric data with the biometric data of the profiles stored in the database, and assign the detected profile ID with the selected marking area,

said at least one processor configured to, while subsequent identification is requested by the application, execute the following operations,
acquiring a second wide image with the camera,
extracting from the second wide image an active area according to the selected marking area,
storing the extracted active area as a second fast scanning area image,
extracting for the active area the biometric data of a face from the second fast scanning area image,
starting the comparison of the extracted biometric data with the biometric data of the profile ID assigned to the selected marking area, and in absence of a match, continuing with the other biometric data of the database until one profile is found,
transmitting the found profile ID to the application.

12. A method to customize an application associated with a television experience based on the recognition of users located in front of a display and in the field of view of a camera, the method comprising:
an initializing operation, during which each user is enrolled in a database of a computer system and is defined by a profile referenced by a profile ID, the profile comprising
a user name,
biometric data of the user acquired by the camera through a face recognition process,
location history information of the user corresponding to the field of view of the camera; and
additional personal data of the user;

a finding operation, during which a wide image, acquired by the camera and embracing a wide area to encompass all potential users located in front of the display, is scanned in order to,
isolate at least one user's face appearing on the wide image,
define a marking area surrounding each isolated face,
store the defined marking area as a fast scanning area image, and
store the position of each of the marking areas in a register that identifies the marking areas;

a matching operation during which each marking area is processed using the stored positions in order to
extract biometric data of the face appearing inside a selected one of the marking areas from the fast scanning area image,
match the extracted biometric data with the biometric data of the profiles stored in the database, and assign the detected profile ID with the selected marking area, and
execute, while subsequent identification is requested by the application, the following operations,
acquiring a second wide image with the camera,
extracting from the second wide image an active area according to the selected marking area,
storing the extracted active area as a second fast scanning area image,
extracting the biometric data of a face from t

Assignees

• Oddou Christophe
• Dagaeff Thierry
• Fishwick Nicholas
• Nagravision Sa

Inventors

• Oddou Christophe
• Dagaeff Thierry
• Fishwick Nicholas

Classifications

• G06V40/16

  Human faces, e.g. facial parts, sketches or expressions · CPC title

• G06V10/143 Primary

  Sensing or illuminating at different wavelengths · CPC title

• G06V10/768

  using context analysis, e.g. recognition aided by known co-occurring patterns · CPC title

• G06V10/245

  by locating a pattern; Special marks for positioning · CPC title

• G06K9/72

  Physics · mapped topic