Method and system for perspectively distoring an image and implanting same into a video stream

• US 5,731,846 A
• Filed: 01/18/1996
• Issued: 03/24/1998
• Est. Priority Date: 03/14/1994
• Status: Expired due to Term

• Current Assignee: PVI Virtual Media Services, LLC (Altice USA, Inc.)
• Original Assignee: SciDel Technologies Ltd. (Gabriel Technologies Corporation)
• Inventors: Tirosh, Ehud, Bar-El, Dan, Amir, Yoel, Kreitman, Haim
• Primary Examiner(s): Peng, John K.
• Assistant Examiner(s): Miller, John W.
• Application Number:
  US08/588,221
• Time in Patent Office: 796 Days
• Field of Search: 348/578, 348/580, 348/581, 348/582, 348/584, 348/585, 348/586, 348/587, 348/590, 348/591, 348/592, 348/596, 348/598, 348/599, 348/722, 345/139
• US Class Current: 348/581
• CPC Class Codes: H04N 5/2628 Alteration of picture size, shape, position or orientation, e.g. zooming, rotation, rolling, perspective, translation
• CPC Class Codes: H04N 5/272 Means for inserting a foreground image in a background image, i.e. inlay, outlay
• CPC Class Codes: H04N 5/2723 Insertion of virtual advertisement; Replacing advertisements physical present in the scene by virtual advertisement data processing systems or methods specially adapted for marketing G06Q30/02

Abstract

A method and system for implanting an image onto a selected one at a time of a plurality of video frames representing a stream of action occurring within a background space, the space having fixed planar surfaces and being scanned by at least one video camera. The method also includes a step of generating a model of a selected one of the fixed surfaces, the model includes a representation of geometrical features characterizing the surface. The model also includes a step of utilizing the model to perspectively distort the image so as to provide it in generally the same perspective of the selected frame, a step of producing a background mask of the selected frame, the frame includes at lest a portion of the fixed surface, the background mask defines changeable regions of the selected frame and unchangeable regions thereof and the step of blending the perspectively distorted image into the selected portion of the changeable region, thereby implanting said image into said selected frame. The system includes means for carrying out the steps of the method.

Claims

1. A method for implanting an image into a selected one at a time of a plurality of video frames representing a stream of action occurring within a background space, the space having fixed planar surfaces and being scanned by at least one video camera, the method comprising:

generating a model, independent of said plurality of video frames, of a selected one of said fixed surfaces, said model comprising a representation of geometrical features characterizing said surface;

utilizing said model to perspectively distort said image so as to provide it in generally the same perspective of said selected frame;

producing a background mask of said selected frame, said frame comprising at least a portion of said fixed surface, said background mask defines changeable regions of said selected frame and unchangeable regions thereof; and

blending said perspectively distorted image into said selected portion of said changeable region, thereby implanting said image into said selected frame.

2. A method according to claim 1, wherein said utilizing compressing utilizing said model and a selected changeable region for producing transformation matrix, said transformation matrix perspectively distort said image.

3. A method according to claim 1, further comprising producing, prior to said utilizing, an image location mask defining the location of said image with respect to said model, thereby said image location mask being distorted with said image.

4. A method according to claim 3, wherein said blending comprising adding said distorted location mask and said selected changeable region of said background mask, thereby producing a permission mask defining pixels in said frame in which corresponding pixels of said image are implanted.

5. A method according to claim 4, further comprising producing a perspectively distorted blending mask, said blending mask defining transparency values and wherein said blending further comprising mixing said image, said permission mask and said blending mask, thereby implanting said image with a desired color value.

6. A method according to claim 5, wherein said mixing comprising assigning a color value to each pixel (x,y), wherein

color value for pixel (x,y)=β

(x,y)*image(x,y)+(1-β

(x,y))*frame(x,y)

wherein β

(x,y)=α

(x,y)*P(x,y)

wherein α

(x,y) and P(x,y) are the blending and permission masks values for the pixel (x,y) and

wherein image (x,y) is said image for the pixel (x,y) and said frame (x,y) are said selected changeable region values for the pixel (x,y).

7. A system for implanting an image into a selected one at a time of a plurality of video frames representing a stream of action occurring within a background space, the space having fixed planar surfaces and being scanned by at least one video camera, comprising:

means for generating a model, independent of said plurality of video frames, of a selected one of said fixed surfaces, said model comprising a representation of geometrical features characterizing said surface;

means for utilizing said model to perspectively distort said image so as to provide it in generally the same perspective of said selected frame;

means for producing a background mask of said selected frame, said frame comprising at least a portion of said fixed surface, said background mask defines changeable regions of said selected frame and unchangeable regions thereof; and

means for blending said perspectively distorted image into said selected portion of said changeable region, thereby implanting said image into said selected frame.

8. A system according to claim 7, wherein said means for utilizing comprising means for utilizing said model and a selected changeable region for producing transformation matrix, said transformation matrix perspectively distort said image.

9. A system according to claim 7, further comprising means for producing, operation prior to said means for utilizing, an image location mask defining the location of said image with respect to said model, thereby said image location mask being distorted with said image.

10. A system according to claim 9, wherein said means for blending comprising means for adding said distorted location mask and said selected changeable region of said background mask, thereby producing a permission mask defining pixels in said frame in which corresponding pixels of said image are implanted.

11. A system according to claim 10, further comprising means for producing a perspectively distorted blending mask, said blending mask defining transparency values and wherein said means for blending further comprising means for mixing said image, said permission mask and said blending mask, thereby implanting said image with a desired color values.

12. A system according to claim 11, wherein said means for mixing comprising means for assigning a color value to each pixel (x,y), wherein

color value for pixel (x,y)=β

(x,y)*image(x,y)+(1-β

(x,y))*frame(x,y)

wherein β

(x,y)=α

(x,y)*P(x,y)

wherein α

(x,y) and P(x,y) are the blending and permission masks values for the pixel (x,y) and

wherein image (x,y) is said image value for the pixel (x,y) and said frame (x,y) are said selected changeable region value for the pixel (x,y).