Systems and methods for image or video personalization with selectable effects
First Claim
1. A method of producing personalized image data for rendering on a display device, comprising:
- accessing image data from an image source, wherein the image data comprises a sequence of image frames;
automatically identifying a key image frame of the sequence of image frames, based on smooth region finding operations, detected regions of a specific texture, detected planar surfaces, detected text, or detected specific motion characters;
receiving, via user input, at least one personalized object for insertion into the image data;
inserting the at least one personalized object into the key image frame to produce personalized output image data wherein at least one of the original image data and the personalized object exhibits motion;
extending the incorporation of the at least one personalized object to additional image frames of the image data;
interpolating motion vectors to determine the location and appearance of the at least one personalized object for the additional image frames;
subsequent to interpolating the motion vectors, querying a user to identify any errors in the location or the appearance of the at least one personalized object for the additional image frames;
in response to the querying, receiving user correction input to adjust the location or the appearance of the at least one personalized object for the additional image frames; and
interpolating motion vectors to determine the location and appearance of the at least one personalized object for the additional image frames based on the user correction input.
7 Assignments
0 Petitions
Accused Products
Abstract
Embodiments relate to systems and methods for image or video personalization with selectable effects. Image data, which can include video sequences or digital still images, can be received in a graphical personalization tool to perform various image processing and related operations to insert personalized objects into the image data. In aspects, the personalized object(s) can be or include graphical inputs such as, for instance, textual information, graphical information, and/or other visual objects. The graphical personalization tool can automatically perform one or more processing stages in the image path, such as identifying key regions in a still image and/or key frames in a video sequence, in which personalized objects will be generated and inserted. Personalized objects can be extended to additional regions of a still image, can be animated across multiple still images, and/or can be extended to additional frames of a video sequence, all on an automated or user-assisted basis.
-
Citations
23 Claims
-
1. A method of producing personalized image data for rendering on a display device, comprising:
-
accessing image data from an image source, wherein the image data comprises a sequence of image frames; automatically identifying a key image frame of the sequence of image frames, based on smooth region finding operations, detected regions of a specific texture, detected planar surfaces, detected text, or detected specific motion characters; receiving, via user input, at least one personalized object for insertion into the image data; inserting the at least one personalized object into the key image frame to produce personalized output image data wherein at least one of the original image data and the personalized object exhibits motion; extending the incorporation of the at least one personalized object to additional image frames of the image data; interpolating motion vectors to determine the location and appearance of the at least one personalized object for the additional image frames; subsequent to interpolating the motion vectors, querying a user to identify any errors in the location or the appearance of the at least one personalized object for the additional image frames; in response to the querying, receiving user correction input to adjust the location or the appearance of the at least one personalized object for the additional image frames; and interpolating motion vectors to determine the location and appearance of the at least one personalized object for the additional image frames based on the user correction input. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11)
-
-
12. A system, comprising:
-
an interface to an image source providing image data; and a processor, communicating with the image source via the interface, the processor being configured to— access the image data from the image source, wherein the image data comprises a sequence of image frames, automatically identify a key image frame of the sequence of image frames, based on smooth region finding operations, detected regions of a specific texture, detected planar surfaces, detected text, or detected specific motion characters, receive, via user input, at least one personalized object for insertion into the image data, insert the at least one personalized object into the key image frame to produce personalized output image data wherein at least one of the original image data and the personalized object exhibits motion, extending the incorporation of the at least one personalized object to additional image frames of the image data; interpolating motion vectors to determine the location and appearance of the at least one personalized object for the additional image frames; subsequent to interpolating the motion vectors, querying a user to identify any errors in the location or the appearance of the at least one personalized object for the additional image frames; in response to the querying, receiving user correction input to adjust the location or the appearance of the at least one personalized object for the additional image frames; and interpolating motion vectors to determine the location and appearance of the at least one personalized object for the additional image frames based on the user correction input. - View Dependent Claims (13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23)
-
Specification