METHOD AND SYSTEM FOR AUTOMATIC CAMERA CONTROL

• US 20090015658A1
• Filed: 07/11/2008
• Published: 01/15/2009
• Est. Priority Date: 07/13/2007
• Status: Active Grant

• Current Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Original Assignee: Tandberg Telecom AS (Cisco Systems, Inc.)
• Inventors: KORNELIUSSEN, Jan Tore, HUSOY, Per Ove, ENSTAD, Gisle
• Granted Patent: US 8,169,463 B2
• Time in Patent Office: Days
• US Class Current: 348/14.80
• CPC Class Codes: H04N 21/4223 Cameras H04N23/00 takes precedence
• CPC Class Codes: H04N 21/478 Supplemental services, e.g. displaying phone caller identification, shopping application
• CPC Class Codes: H04N 21/4788 communicating with other users, e.g. chatting arrangements for providing for computer conferences, e.g. chat rooms, to substation in data switching networks H04L12/1813; distributed application using peer-to-peer [P2P] networks H04L67/104
• CPC Class Codes: H04N 23/611 where the recognised objects include parts of the human body
• CPC Class Codes: H04N 23/69 Control of means for changing angle of the field of view, e.g. optical zoom objectives or electronic zooming
• CPC Class Codes: H04N 7/142 Constructional details of the terminal equipment, e.g. arrangements of the camera and the display
• CPC Class Codes: H04N 7/15 Conference systems

Abstract

A method for automatically determining an orientation and zoom of an image pickup device associated with a video conferencing system, wherein the method includes the steps of: generating, at the image pickup device, an image signal representative of an image framed by the image pickup device; processing the image signal to identify objects plural users of the video conferencing system in the image; steering the image pickup device to an initial orientation; determining a location of all the identified objects relative to a reference point and determining respective sizes of the identified objects; defining an area of interest in the image, wherein the area of interest includes all the identified objects; and steering the image pickup device to frame the defined area of interest including all the identified objects where a center of the frame substantially coincides with a center of all the identified objects.

Claims

1. A method for automatically determining an orientation and zoom of an image pickup device associated with a video conferencing system, wherein said method comprises the steps of:

generating, at said image pickup device, an image signal representative of an image framed by said image pickup device;

processing the image signal to identify objects in said image;

steering the image pickup device to an initial orientation;

determining a location of all the identified objects relative to a reference point and determining respective sizes of the identified objects;

defining an area of interest in said image, wherein said area of interest includes all the identified objects; and

steering the image pickup device to frame said defined area of interest including all the identified objects where a center of the frame substantially coincides with a center formed by a group of all the identified objects.

2. The method according to claim 1, where said step of steering the image pickup device to frame said defined area of interest comprises:

varying an azimuth angle and an elevation angle of the image pickup device; and

varying a zoom of the image pickup device.

3. The method according to claim 1, where said step of steering the image pickup device to an initial orientation further comprise:

zooming the image pickup device out to a maximum field of view;

orienting the image pickup device according to a predefined pan- and tilt-orientation so as to frame as much as possible of a room in which the image pickup device is situated to capture.

4. The method according to claim 1, wherein the image signal represents a frame of a video, and the step of processing the image signal to identify objects further comprises:

detecting images of faces or heads that appear in a series of frames of the video;

tracking the detected faces or heads through the series of frames of the video;

registering a detection of the face or head if said detection occurs in a predefined number of frames in the series of frames.

5. The method according to claim 4, wherein the step of defining the area of interest further comprises:

defining a set of margins for a first area, where said first area is a smallest definable area enclosing all of said registered images of faces or heads; and

defining said area of interest by expanding said margins by a set of offset values.

6. The method according to claim 5, wherein the area of interest is further defined to fit a predetermined image aspect ratio.

7. The method according to claim 1, wherein said area of interest represents a close-up view of the objects.

8. The method according to claim 1, wherein the method is implemented in response to receiving an indication of a predefined event, wherein the predefined event includes one of:

switching on a video conferencing system, receiving or sending a conference call initiation request, and receiving a command from a user.

9. The method according to claim 1, where said method further comprises:

processing an audio signal from a set of audio pickup devices to determine a location of an audio source relative to a reference point.

10. The method according to claim 1, wherein said method is implemented in response to detecting a presence of an audio source outside the framed area of interest.

11. The method according to claim 1, wherein said method is implemented in response to detecting a disappearance of one or more of the objects from the framed area of interest.

12. The method of claim 1, wherein the steering the image pickup device to the initial orientation includes using digital pan, tilt, or zoom, and the steering the image pickup device to frame said defined area of interest includes using digital pan, tilt, or zoom.

13. The method of claim 1, wherein the method further comprises:

comparing a vertical length of the area of interest to a horizontal length of the area of interest, wherein each of the vertical length and the horizontal length relate to a zoom factor; and

adjusting the image pickup device to have a degree of zoom corresponding with a smallest of the vertical length and the horizontal length.

14. A system for automatically determining an orientation and zoom of an image pickup device associated with a video conferencing system, wherein said system comprises:

an image pickup device configured to generate image signals representative of an image framed by said image pickup device;

a video detection unit configured to process the image signal to identify objects in said image, to determine a location of all the identified objects relative to a reference point, and to determine respective sizes of the identified objects;

an image processing unit configured to define an area of interest in said image, wherein said area of interest includes all the identified objects; and

a control unit configured to, on an occurrence of a predefined event,

steer the image pickup device to an initial orientation, to receive camera coordinates from said image processing unit corresponding to said area of interest, to steer the image pickup device to frame said area of interest including all the identified objects where a center of the frame substantially coincides with a center formed by a group of all the identified objects.

15. The system according to claim 14, wherein the image signal represents a frame of a video, and where the identified objects are detected images of faces or heads that appear in a series of frames of the video.

16. The system according to claim 15, further comprising:

a face tracking unit configured to track the detected faces or heads through the series of frames, and to register a detection as a face or head if said detection occurs in a predefined number of frames in the series of frames.

17. The system according to claim 15, wherein said image processing unit is further configured to define a set of margins for a first area, where said first area is a smallest definable area enclosing all of said registered images of faces or heads, and to define said area of interest by expanding said margins by a set of offset values.

18. The system according to claim 17, where the area of interest is defined to fit a predetermined image aspect ratio.

19. The system according to claim 14, further comprising:

an audio source locator configured to process an audio signal from a set of audio pickup devices, to determine a location of an audio source in camera coordinates.

20. The system according to claim 14, wherein said control unit is further configured to receive audio source coordinates from said audio source locator, and to compare said audio source coordinates to a current field of view.

21. The system according to claim 14, wherein said image pickup device comprises a camera control unit for positioning said image pickup device, wherein the control unit is configured to supply control signals to said camera control unit for the orientation and zoom of the image pickup device, the control signal being generated based on said area of interest.

22. The system according to claim 14, wherein the predefined event comprises detecting a presence of an audio source outside a current field of view.

23. The system according to claim 14, wherein the predefined event comprises detecting a disappearance of one or more of the objects from the framed area of interest.

24. The system according to claim 14, where the said predefined event comprises detecting a presence of an audio source outside the framed area of interest.

25. The system of claim 14, wherein the image processing unit is further configured to compare a vertical length of the area of interest to a horizontal length of the area of interest, wherein each of the vertical length and the horizontal length relate to a zoom factor, and to adjust the image pickup device to have a degree of zoom corresponding with a smallest of the vertical length and the horizontal length.

26. A computer readable storage medium encoded with instruction, which when executed by a computer, causes the computer to implement a method for automatically determining an orientation and zoom of an image pickup device associated with a video conferencing system, wherein said method comprises the steps of:

generating, at said image pickup device, an image signal representative of an image framed by said image pickup device;

processing the image signal to identify objects plural users of the video conferencing system in said image;

steering the image pickup device to an initial orientation;

determining a location of all the identified objects relative to a reference point and determining respective sizes of the identified objects;

defining an area of interest in said image, wherein said area of interest includes all the identified objects; and

steering the image pickup device to frame said defined area of interest including all the identified objects where a center of the frame substantially coincides with a center formed by a group of all the identified objects.

27. A system for automatically determining an orientation and zoom of an image pickup device associated with a video conferencing system, comprising:

means for generating, at said image pickup device, an image signal representative of an image framed by said image pickup device;

means for processing the image signal to identify objects plural users of the video conferencing system in said image;

means for steering the image pickup device to an initial orientation;

means for determining a location of all the identified objects relative to a reference point and determining respective sizes of the identified objects;

means for defining an area of interest in said image, wherein said area of interest includes all the identified objects; and

means for steering the image pickup device to frame said defined area of interest including all the identified objects where a center of the frame substantially coincides with a center formed by a group of all the identified objects.