Method of Tracking the Position of the Head in Real Time in a Video Image Stream

• US 20090129631A1
• Filed: 01/31/2007
• Published: 05/21/2009
• Est. Priority Date: 02/07/2006
• Status: Active Grant

• Current Assignee: Orange S.A.
• Original Assignee: Orange S.A.
• Inventors: Pallone, Gregory, Faure, Julien
• Granted Patent: US 8,571,258 B2
• Time in Patent Office: Days
• US Class Current: 382/103
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/30196 Human being; Person
• CPC Class Codes: G06T 7/248 involving reference images or patches
• CPC Class Codes: G06T 7/73 using feature-based methods
• CPC Class Codes: G06V 40/167 using comparisons between temporally consecutive images

Abstract

The invention relates to a method of tracking the position of the bust of a user on the basis of a video image stream, said bus comprising the user'"'"'s torso and head, the method comprising the determination of the position of the torso on a first image, in which method a virtual reference frame is associated with the torso on said first image, and in which method, for a second image, a new position of the virtual reference frame is determined on said second image, and, a relative position of the head with respect to said new position of the virtual reference frame is measured by comparison with the position of the virtual reference frame on said first image, so as to determine independently the movements of the head and the torso.

Claims

1. A method of tracking the position of the upper body of a user from a video image stream, said upper body comprising the torso and the head of the user, the method comprising a determination of the position of the torso (T_x0, T_y0) in a first image, wherein:

a virtual reference frame is associated with the torso in said first image;

and in that, for a second image;

a new position of the virtual reference frame is determined in said second image; and

a relative position of the head (El, Az, P_x, P_y) relative to said new position of the virtual reference frame is measured by comparison with the position of the virtual reference frame in said first image, in order to determine independently the movements of the head and of the torso.

2. The method as claimed in claim 1, wherein the head of the user comprises the face of the user, and wherein:

at least one position of the face (V_x, V_y) is measured relative to the new position of the reference frame in the second image in order to deduce from this the relative position of the head.

3. The method as claimed in claim 1, wherein in order to associate a virtual reference frame with the first image, the position of the face (V_x0, V_y0) of the user is determined in said first image, and from this the position of the torso (T_x0, Ty₀) in said first image is deduced.

4. The method as claimed in claim 3, wherein the position of the torso (Tx₀, Ty₀) is determined vertically from the position of the face.

5. The method as claimed in claim 2, wherein the relative position of the head is measured from a comparison between the positions of the torso (T₀, T_y0) and of the face (V₀, V_y0) in the first image, and the new positions of the torso (T_x, T_y) and of the face (Vx, Vy) in said second image.

6. The method as claimed in claim 1, wherein the head includes the neck (135), and wherein the relative position of the head is characterized by:

an elevation corresponding to the rotation of the head between the first and the second image, measured in the plane of symmetry of said head;

an azimuth corresponding to the rotation of the head between the first and the second image, measured about an axis defined by the neck of the user; and

a translation of the head between the first and the second image.

7. The method as claimed in claim 6, wherein the relative position of the head is given by the expressions:

$P_x=T_x-T_{x<br><br><br><br>0}$ $P_y=T_y-T_{y<br><br><br><br>0}$ $E<br><br><br><br>I=\frac{\mathrm{\pi <br><br>}}{2}-\arccos <br><br>\left(\left[\left(V_{y<br><br><br><br>0}-T_{y<br><br><br><br>0}\right)-\left(V_y-T_y\right)\right]·<br><br>C_1^{-1}\right)$ $\mathrm{Az}=\frac{\mathrm{\pi <br><br>}}{2}-\arccos <br><br>\left(\left[\frac{\left(V_{x<br><br><br><br>0}-T_{x<br><br><br><br>0}\right)-\left(V_x-T_x\right)}{\cos <br><br>\left(E<br><br><br><br>I\right)}\right]·<br><br>C_2^{-1}\right)$

with

T_x0 and T_y0 the coordinates representing the position of the torso in the first image;

T_x and T_y the coordinates representing the position of the torso in the second image;

V_x0 and V_y0 the coordinates representing the position of the face in the first image;

V_x and V_y the coordinates representing the position of the face in the second image;

these coordinates being measured in a shared reference frame linked with the horizontal and vertical axes of the first and second images; and

C1 and C₂ are constants such that;

C₁>

[(V_y0−

T_y0)−

(V_y−

T_y)]

$C_2><br><br>\left[\frac{\left(V_{x<br><br><br><br>0}-T_{x<br><br><br><br>0}\right)-\left(V_x-T_x\right)}{\cos <br><br>\left(E<br><br><br><br>I\right)}\right]<br><br><br><br>\mathrm{and}-\frac{\mathrm{\pi <br><br>}}{2}<<br><br>E<br><br><br><br>I<<br><br>\frac{\mathrm{\pi <br><br>}}{2}$

with

P_x the horizontal component of the translation of the head;

P_y the vertical component of the translation of the head;

El the elevation; and

Az the azimuth.

8. The method as claimed in claim 7, wherein the coordinates representing the position of the face and of the torso in the first image are the coordinates of the center of gravity of a first cluster of points defined on the face and the coordinates of the center of gravity of a second cluster of points defined on the torso, the coordinates representing the position of the face and of the torso in the second image being the coordinates of the centers of gravity of the first and second clusters of points considered in their new position in said second image.

9. The method as claimed in claim 1, wherein the first image is a calibration image and the second image is a current image from the video image stream.

10. A system for tracking the position of the upper body of a user from a video image stream, the upper body comprising the head and the torso of the user, the system comprising:

a video-image capture system for generating at least one first image and a second image of the upper body of the user;

a memory; and

a processing unit connected with said video-image capture system, said processing unit making use of instructions available in said memory to implement the method according to claim 1.

11. A computer program product designed to be executed in a memory of the processing unit of a computer system, the program comprising instructions for the implementation of the method according to claim 1 when it is executed in the processing unit.