Back to results
US 9,179,078 B2Granted
Combining multiple video streams
AssigneeHewlett-Packard Development Company, L.P.
InventorsRamin Samadani, Kar-Han Tan
Life of this patent
Estimating the term…
Abstract
Methods, computer-readable media, and systems are provided for combining multiple video streams. One method for combining the multiple video streams includes extracting a sequence of media frames (224-1/224-2) from presenter (222-1) video and from shared digital rich media (222-2) video (340). The media frame (224-1/224-2) content is analyzed (226) to determine a set of space and time varying alpha values (228/342). A compositing operation (230) is performed to produce the combined video frames (232) based on the content analysis (226/344).
Claims14
A method comprising:
- capturing presenter video of a presenter, the presenter video temporally divided into a sequence of media frames, each media frame divided into spatial regions;
- by a processor:
- extracting the sequence from the video;
- analyzing the content on the media frames to determine a set of space and time varying alpha values, wherein the alpha values temporally vary in corresponding spatial regions over the sequence and spatially vary in each media frame over the spatial regions thereof,
- by assigning each region of each media frame a distinct alpha value equal to:
- a first value when the region includes presenter video background and a corresponding region of a corresponding media frame of shared digital rich media does not include text, or when the region does not includes a face or facial features of the presenter and the corresponding region includes the text;
- a second value less than the first value when the region includes the face or the facial features and the corresponding region does not include the text;
- or
- a third value less than the first value and greater than the second value when the region include the face or the facial features of the presenter and the corresponding includes the text;
- performing a compositing operation on the media frames in relation to corresponding media frames of the shared digital rich media to produce combined video frames based on alpha values of the regions of the media frames;
- and
- displaying a video of the combined video frames.
9 more elements
A computer-readable non-transitory medium storing a set of instructions executable by the computer to cause the computer to:
- receive presenter video of a presenter, the presenter video temporally divided into a sequence of media frames, each media frame divided into spatial regions;
- extract the sequence from the video;
- analyze the content on the media frames to determine a set of space and time varying alpha values, wherein the alpha values temporally vary in corresponding spatial regions over the sequence and spatially vary in each media frame over the spatial regions thereof, by assigning each region of each media frame a distinct alpha value equal to:
- a first value when the region includes presenter video background and a corresponding region of a corresponding media frame of shared digital rich media does not include text, or when the region does not includes a face or facial features of the presenter and the corresponding region includes the text;
- a second value less than the first value when the region includes the face or the facial features and the corresponding region does not include the text;
- or
- a third value less than the first value and greater than the second value when the region include the face or the facial features of the presenter and the corresponding includes the text;
- perform a compositing operation on the media frames in relation to corresponding media frames of the shared digital rich media to produce combined video frames based on alpha values of the regions of the media frames;
- and
- display a video of the combined video frames.
7 more elements
A presentation system comprising:
- a processor and a memory coupled to the processor, wherein the memory includes stored executable instructions executed by the processor to:
- receive presenter video of a presenter, the presenter video temporally divided into a sequence of media frames, each media frame divided into spatial regions;
- extract a sequence of media frames from the video;
- analyze the content on the media frames to determine a set of space and time varying alpha values, wherein the alpha values temporally vary in corresponding spatial regions over the sequence and spatially vary in each media frame over the spatial regions thereof, by assigning each region of each media frame a distinct alpha value equal to:
- a first value when the region includes presenter video background and a corresponding region of a corresponding media frame of shared digital rich media does not include text, or when the region does not includes a face or facial features of the presenter and the corresponding region includes the text;
- a second value less than the first value when the region includes the face or the facial features and the corresponding region does not include the text;
- or
- a third value less than the first value and greater than the second value when the region include the face or the facial features of the presenter and the corresponding includes the text;
- perform a compositing operation on the media frames in relation to corresponding media frames of the shared digital rich media to produce combined video frames based on alpha values of the regions of the media frames;
- and
- cause a video of the combined video frames to be displayed.
8 more elements
Dependencies are inferred from the claim text.