Automatic compositing of 3D objects in a still frame or series of frames
First Claim
Patent Images
1. A method for generating an image, comprising the computer implemented steps of:
- analyzing a two dimensional (2D) image for determining several properties thereof;
detecting edges in the 2D image based on the properties of the 2D image;
finding objects in the 2D image based at least in part on the properties of the 2D image;
adding a three dimensional (3D) object to the 2D image;
performing a depth analysis for estimating which objects in the 2D image are positioned in front of the 3D object, or which objects in the 2D image are positioned behind the 3D object, wherein the depth analysis includes analyzing vertical positions of at least portions of the 2D and 3D objects relative to each other and estimating which objects in the 2D image are positioned in front of or behind the 3D object based on the vertical positions of the at least portions of the 2D and 3D objects; and
rendering at least one of the following effects;
a reflection of one of the objects in the 2D image on an outer surface of the 3D object, a representation of one of the objects in the 2D image viewable through a transparent or semi-transparent portion of the 3D object, and an effect of refraction on one of the objects in the 2D image viewable through a transparent or semi-transparent portion of the 3D object; and
rendering a shadow cast by one of the objects in the 2D image on an outer surface of the 3D object.
3 Assignments
0 Petitions
Accused Products
Abstract
Systems, methods, and computer program products for creating composite images. In one approach, a method for generating an image, comprises the computer implemented steps of: adding a three dimensional (3D) object to a two dimensional (2D) image; rendering a representation of one of the objects in the 2D image viewable through a transparent or semi-transparent portion of the 3D object; and rendering an effect of refraction on one of the objects in the 2D image viewable through the transparent or semi-transparent portion of the 3D object. Additional embodiments are also presented.
38 Citations
27 Claims
-
1. A method for generating an image, comprising the computer implemented steps of:
-
analyzing a two dimensional (2D) image for determining several properties thereof; detecting edges in the 2D image based on the properties of the 2D image; finding objects in the 2D image based at least in part on the properties of the 2D image; adding a three dimensional (3D) object to the 2D image; performing a depth analysis for estimating which objects in the 2D image are positioned in front of the 3D object, or which objects in the 2D image are positioned behind the 3D object, wherein the depth analysis includes analyzing vertical positions of at least portions of the 2D and 3D objects relative to each other and estimating which objects in the 2D image are positioned in front of or behind the 3D object based on the vertical positions of the at least portions of the 2D and 3D objects; and rendering at least one of the following effects;
a reflection of one of the objects in the 2D image on an outer surface of the 3D object, a representation of one of the objects in the 2D image viewable through a transparent or semi-transparent portion of the 3D object, and an effect of refraction on one of the objects in the 2D image viewable through a transparent or semi-transparent portion of the 3D object; andrendering a shadow cast by one of the objects in the 2D image on an outer surface of the 3D object. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19)
-
-
20. A method for generating an image, comprising the computer implemented steps of:
-
analyzing a two dimensional (2D) image for determining several properties thereof; detecting edges in the 2D image based on the properties of the 2D image; refining lines based on the detected edges in the 2D image; detecting shadows in the 2D image based on the properties of the 2D image; adjusting a boundary of the detected shadows in the 2D image based on a proximity to the detected edges; finding objects in the 2D image based at least in part on the properties of the 2D image; adding a three dimensional (3D) object to the 2D image; performing a depth analysis for estimating which objects in the 2D image are positioned in front of the 3D object, or which objects in the 2D image are positioned behind the 3D object, wherein the depth analysis includes analyzing vertical positions of at least portions of the 2D and 3D objects relative to each other and estimating which objects in the 2D image are positioned in front of or behind the 3D object based on the vertical positions of the at least portions of the 2D and 3D objects; rendering at least one of the following effects;
a reflection of one of the objects in the 2D image on an outer surface of the 3D object, a shadow cast by one of the objects in the 2D image on an outer surface of the 3D object, a representation of one of the objects in the 2D image viewable through a transparent or semi-transparent portion of the 3D object, and an effect of refraction on one of the objects in the 2D image viewable through a transparent or semi-transparent portion of the 3D object; andallowing a user to apply masks for manipulating the image. - View Dependent Claims (21, 22)
-
-
23. A method for generating an image, comprising the computer implemented steps of:
-
analyzing a two dimensional (2D) image for determining several properties thereof; adding a three dimensional (3D) object to the 2D image; finding objects in the 2D image based at least in part on the properties of the 2D image; performing a depth analysis for estimating which objects in the 2D image are positioned in front of the 3D object, or which objects in the 2D image are positioned behind the 3D object, wherein the depth analysis includes analyzing vertical positions of at least portions of the 2D and 3D objects relative to each other and estimating which objects in the 2D image are positioned in front of or behind the 3D object based on the vertical positions of the at least portions of the 2D and 3D objects; rendering a representation of one of the objects in the 2D image viewable through a transparent or semi-transparent portion of the 3D object; and rendering an effect of refraction on one of the objects in the 2D image viewable through the transparent or semi-transparent portion of the 3D object; performing radiosity processing; performing caustics processing; and adding motion blur to the 3D object, wherein the objects in the 2D image are found using at least one of a histogram, motion-based detection and using leveling data, wherein the depth analysis includes analyzing vertical positions of at least portions of the objects relative to each other. - View Dependent Claims (24)
-
-
25. A system for generating an image, comprising:
-
hardware for analyzing a two dimensional (2D) image for determining several properties thereof; hardware for adding a three dimensional (3D) object to the 2D image; hardware for finding objects in the 2D image based at least in part on the properties of the 2D image, wherein the objects in the 2D image are found using at least one of a histogram, motion-based detection and using leveling data; hardware for detecting edges in the 2D image based on the properties of the 2D image; hardware for refining lines based on the detected edges in the 2D image; hardware for detecting shadows in the 2D image based on the properties of the 2D image; hardware for adjusting a boundary of the detected shadows in the 2D image based on a proximity to the detected edges; hardware for performing a depth analysis for estimating which objects in the 2D image are positioned in front of the 3D object, or which objects in the 2D image are positioned behind the 3D object, wherein the depth analysis includes analyzing vertical positions of at least portions of the 2D and 3D objects relative to each other and estimating which objects in the 2D image are positioned in front of or behind the 3D object based on the vertical positions of the at least portions of the 2D and 3D objects, wherein the depth analysis does the estimating without Z-data of the 2D image; hardware for rendering a reflection of one of the objects in the 2D image on an outer surface of the 3D object; hardware for rendering a representation of one of the objects in the 2D image viewable through a transparent or semi-transparent portion of the 3D object; and hardware for rendering an effect of refraction on one of the objects in the 2D image viewable through the transparent or semi-transparent portion of the 3D object; hardware for performing radiosity processing; hardware for performing caustics processing; hardware for rendering a shadow cast by one of the objects in the 2D image on an outer surface of the 3D object; and hardware for adding motion blur to the 3D object, wherein the objects in the 2D image are found using at least one of a histogram, motion-based detection and using leveling data.
-
-
26. A computer program product, comprising:
-
a non-transitory computer readable medium having computer code thereon, the computer code including; computer code for analyzing a two dimensional (2D) image for determining several properties thereof; computer code for adding a three dimensional (3D) object to the 2D image; computer code for finding objects in the 2D image based at least in part on the properties of the 2D image; computer code for performing a depth analysis for estimating which objects in the 2D image are positioned in front of the 3D object, or which objects in the 2D image are positioned behind the 3D object, wherein the depth analysis includes analyzing vertical positions of at least portions of the 2D and 3D objects relative to each other and estimating which objects in the 2D image are positioned in front of or behind the 3D object based on the vertical positions of the at least portions of the 2D and 3D objects; computer code for rendering a reflection of one of the objects in the 2D image on an outer surface of the 3D object; computer code for rendering a representation of one of the objects in the 2D image viewable through a transparent or semi-transparent portion of the 3D object; and computer code for rendering an effect of refraction on one of the objects in the 2D image viewable through the transparent or semi-transparent portion of the 3D object; computer code for performing radiosity processing; computer code for performing caustics processing; computer code for rendering a shadow cast by one of the objects in the 2D image on an outer surface of the 3D object; and computer code for adding motion blur to the 3D object, wherein the objects in the 2D image are found using at least one of a histogram, motion-based detection and using leveling data. - View Dependent Claims (27)
-
Specification