- Number
- 10264273
- Published
- 2019-04-16
- Filed
- 2014-10-31
- Assignee
- Disney Enterprises, Inc.
- Inventors
- Farré Guiu; Miquel À.
- CPC
- H04L65/70; H04N19/40
- Verdict
- Set aside computed information metadata extraction applied to transcoding, codec plumbing
- Source
- Google Patents · FreePatentsOnline
Abstract
Systems, methods and articles of manufacture for transcoding video content. Embodiments include receiving an instance of video content for processing. A plurality of shots within the instance of video content is determined. Embodiments analyze the instance of video content to generate metadata describing the media content. The generated metadata includes, for each of the plurality of shots, data describing a plurality of frames within the respective shot. An optimized transcoding schedule for transcoding the instance of video content from a first video encoding format to a second video encoding format is generated based on the generated metadata. Embodiments further include transcoding the instance of video content according to the optimized transcoding schedule.
Background
BACKGROUND(1) Field of the Invention(2) The present disclosure relates to transcoding media content, and more specifically, to techniques for extracting metadata from video content as part of a transcoding operation and using the extracted metadata to optimize the transcoding operation for the video content.(3) Description of the Related Art(4) A number of different techniques exist today for delivering video content to users. Generally speaking, existing systems for delivering video content, such as over-the-air broadcasts, cable television service, Internet television service, telephone network television service, satellite television service, satellite radio service, websites, etc., provide a relatively impersonalized, generic experience to all viewers. For example, with respect to broadcast television, all viewers of a given television network station receive essentially the same content in essentially the same order.(5) As the number of delivery mechanisms for delivering video content to users continues to grow, so does the need to encode the video content in various different formats and bit rates. For example, the encoding format for the video content used in broadcast television may differ from the encoding format used in high-quality media streaming. Moreover, in order to provide an improved media streaming environment, content servers may store the video content in multiple different encoding rates, so that an appropriate encoding rate can be selected for streaming
Claims
1. A computer-implemented method of video encoding based on measures of motion, the computer-implemented method comprising: receiving an instance of video content for processing; determining a plurality of shots within the instance of video content, each shot having a respective plurality of frames; processing the instance of video using one or more optical character recognition (OCR) algorithms in order to generate first OCR results; generating a plurality of measures of motion including, for each of the plurality of shots, a respective measure of motion within the plurality of frames of the respective shot; generating, by operation of one or more computer processors, an optimized transcoding schedule for transcoding the instance of video content from a first video encoding format to a second video encoding format, wherein the optimized transcoding schedule specifies, for each of the plurality of shots, a respective motion search range specifying a size of an area in which a video codec is to search for an element that moved from a previous frame of the plurality of frames of the respective shot, wherein the respective motion search range is determined based on the respective measure of motion; encoding the instance of video content according to the optimized transcoding schedule, wherein the encoded instance of video content has an associated measure of encoding quality higher than in an absence of the measures of motion, while adhering to a file size restriction at least as stringent as in the absence of the measures of motion, wherein the encoded instance of video content is output; processing the encoded instance of video content using the one or more OCR algorithms in order to generate second OCR results; and evaluating a level of refinement of the optimized transcoding schedule by comparing the first OCR results and the second OCR results.
6. A system of video encoding based on measures of motion, the system comprising: one or more computer processors; and a memory containing a program that, when executed by the one or more computer processors, performs an operation comprising: receiving an instance of video content for processing; determining a plurality of shots within the instance of video content, each shot having a respective plurality of frames; processing the instance of video using one or more optical character recognition (OCR) algorithms in order to generate first OCR results; generating a plurality of measures of motion including, for each of the plurality of shots, a respective measure of motion within the plurality of frames of the respective shot; generating an optimized transcoding schedule for transcoding the instance of video content from a first video encoding format to a second video encoding format, wherein the optimized transcoding schedule specifies, for each of the plurality of shots, a respective motion search range specifying a size of an area in which a video codec is to search for an element that moved from a previous frame of the plurality of frames of the respective shot, wherein the respective motion search range is determined based on the respective measure of motion; encoding the instance of video content according to the optimized transcoding schedule, wherein the encoded instance of video content has an associated measure of encoding quality higher than in an absence of the measures of motion, while adhering to a file size restriction at least as stringent as in the absence of the measures of motion, wherein the encoded instance of video content is output; processing the encoded instance of video content using the one or more OCR algorithms in order to generate second OCR results; and evaluating a level of refinement of the optimized transcoding schedule by comparing the first OCR results and the second OCR results.
11. A non-transitory computer-readable medium containing a program that, when executed, performs an operation for video encoding based on measures of motion, the operation comprising: receiving an instance of video content for processing; determining a plurality of shots within the instance of video content, each shot having a respective plurality of frames; processing the instance of video using one or more optical character recognition (OCR) algorithms in order to generate first OCR results; a plurality of measures of motion including, for each of the plurality of shots, a respective measure of motion within the plurality of frames of the respective shot; generating, by operation of one or more computer processors when executing the program, an optimized transcoding schedule for transcoding the instance of video content from a first video encoding format to a second video encoding format, wherein the optimized transcoding schedule specifies, for each of the plurality of shots, a respective motion search range specifying a size of an area in which a video codec is to search for an element that moved from a previous frame of the plurality of frames of the respective shot, wherein the respective motion search range is determined based on the respective measure of motion; encoding the instance of video content according to the optimized transcoding schedule, wherein the encoded instance of video content has an associated measure of encoding quality higher than in an absence of the measures of motion, while adhering to a file size restriction at least as stringent as in the absence of the measures of motion, wherein the encoded instance of video content is output; processing the encoded instance of video content using the one or more OCR algorithms in order to generate second OCR results; and evaluating a level of refinement of the optimized transcoding schedule by comparing the first OCR results and the second OCR results.