Methods, Systems, Devices and Computer Program Products for Adapting External Content to a Video Stream
AssigneeDolby International AB
InventorAntonio MATEOS SOLE
Life of this patent
Estimating the term…
Abstract
This disclosure falls into the field of adapting external content to a video stream, and more specifically it is related to analyzing the video stream to define a suitable narrative model, and adapting the external content based on this narrative model.
Claims19
A method for adapting external content to a video stream having a time span, the video stream comprising a plurality of frames each comprising pixels, the method comprising the steps of:
- computing a first plurality of difference indicators, each difference indicator of the first plurality of difference indicators indicating a difference between values of the pixels between a frame of the plurality of frames and an nth subsequent frame of the plurality of frames, wherein n is an integer≥1;
- computing a second plurality of difference indicators by convolving the first plurality of difference indicators with a strictly decreasing function, each difference indicator of the second plurality of difference indicators corresponding to a specific position in time of the video stream,
- fitting a predetermined parametrizable function to the second plurality of difference indicators, thereby yielding a function descriptive of a narrative model, FNM, for the video stream during the time span of the video stream,
- using the FNM to adapt external content to the video stream.
A system for adapting external content to a video stream having a time span, the video stream comprising a plurality of frames each comprising pixels, the system comprising a device connected to a database, the database comprising external content, wherein the device comprises a processor adapted for:
- computing a first plurality of difference indicators, each difference indicator of the first plurality of difference indicators indicating a difference between values of the pixels between a frame of the plurality of frames and an nth of subsequent frame of the plurality of frames, wherein n is an integer≥1;
- computing a second plurality of difference indicators by convolving the first plurality of difference indicators with a strictly decreasing function, each difference indicator of the second plurality of difference indicators corresponding to a specific position in time of the video stream,
- fitting a predetermined parametrizable function to the second plurality of difference indicator, thereby yielding a function descriptive of a narrative model, FNM, for the video stream during the time span of the video stream,
- selecting external content from the database, and
- using the FNM to adapt the selected external content to the video stream.
Dependencies are inferred from the claim text.