US 11,748,404 B1
Computer video analytics processor
R J Burnham, Winter Park, FL (US); Daniel Evers, Oviedo, FL (US); Markus Hahn, Santa Cruz, CA (US); Syed Zain Masood, Orlando, FL (US); Brent Richardson, Winter Park, FL (US); and Jonathan Taylor, Maitland, FL (US)
Assigned to Sighthound, Inc., Winter Park, FL (US)
Filed by Sighthound, Inc., Winter Park, FL (US)
Filed on Jun. 16, 2020, as Appl. No. 16/903,284.
Claims priority of provisional application 62/862,572, filed on Jun. 17, 2019.
Int. Cl. G06F 16/00 (2019.01); G06F 16/732 (2019.01); G06F 16/783 (2019.01); G06F 16/75 (2019.01); G06F 16/78 (2019.01)
CPC G06F 16/7335 (2019.01) [G06F 16/75 (2019.01); G06F 16/7834 (2019.01); G06F 16/7867 (2019.01)] 27 Claims
OG exemplary drawing
 
1. A method of synchronizing two or more video streams, comprising:
obtaining a plurality of video streams;
using an analytics processor having a correlation stage to identify similarities among the video streams of the plurality of video streams;
distill cross-populated information that associate at least one frame from at least two video streams and relate to location and/or time of video capture of content of the at least two video streams;
compute confidence degrees for the cross-populated information; and
process a mesh of cross-populated information to increase confidence degrees of the cross-populated information,
wherein using the analytics processor to identify similarities among the video streams comprises:
representing references to video inputs using a markup language; and
representing actions to be taken on the video inputs,
wherein the actions are represented by data stored in a data object in the markup language.