US 12,170,814 B2
Feature and parameter extraction for audio and video processing
Wade Wan, Irvine, CA (US); Jason W. Herrick, San Jose, CA (US); Richard Hayden Wyman, San Jose, CA (US); Brett J. Andrews, Irvine, CA (US); and Gheorghe Berbecel, Irvine, CA (US)
Assigned to Avago Technologies International Sales Pte. Limited, Singapore (SG)
Filed by AVAGO TECHNOLOGIES INTERNATIONAL SALES PTE. LIMITED, Singapore (SG)
Filed on Oct. 28, 2021, as Appl. No. 17/513,518.
Prior Publication US 2023/0133015 A1, May 4, 2023
Int. Cl. H04N 21/44 (2011.01); G06N 20/00 (2019.01); H04N 19/14 (2014.01); H04N 19/517 (2014.01); H04N 21/439 (2011.01)
CPC H04N 21/44008 (2013.01) [G06N 20/00 (2019.01); H04N 19/14 (2014.11); H04N 19/517 (2014.11); H04N 21/4394 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method of improving security of media streams, comprising:
receiving, by one or more processors of a device, a decoded media stream from a media decoding pipeline that receives and decodes an encoded media stream;
identifying, by the one or more processors, based on the decoded media stream, a set of features to generate;
generating, by the one or more processors, the set of features using the decoded media
stream, the one or more processors storing the set of features in a first region of memory accessible by a processor executing a machine-learning model while storing the decoded media stream in a second region of memory for which the processor executing the machine-learning model is prevented from accessing; and
providing, by the one or more processors using the first region of memory, the set of features to the processor executing the machine-learning model, wherein the processor executing the machine-learning model is prevented from accessing the second region of memory storing the decoded media stream by the device.