System and method to estimate end-to-end video frame delays

Abstract

System and method to calculate video frame delay in a video stream received by a telecommunications endpoint, the method including: locating reference features characteristic of content in the received video stream; calculating, by use of a processor, reduced reference features from the located reference features; receiving reduced reference features of a transmitted video stream, the transmitted video stream corresponding to the received video stream; calculating, by use of a processor, a received trajectory of the reduced reference features from the received video stream; calculating, by use of a processor, a transmitted trajectory of the reduced reference features from the transmitted video stream; and calculating, by use of a processor, video frame delay as a time shift between the received trajectory and the transmitted trajectory.

What is claimed is: 
     
         1 . A method to calculate video frame delay in a video stream received by a telecommunications endpoint, the method comprising:
 locating, by a processor, reference features that are characteristic of content in the received video stream;   calculating, by the processor, reduced reference features from the located reference features;   receiving, by the processor, reduced reference features of a transmitted video stream, the transmitted video stream corresponding to the received video stream;   calculating, by the processor, a received trajectory of the reduced reference features from the received video stream;   calculating, by the processor, a transmitted trajectory of the reduced reference features from the transmitted video stream; and   calculating, by the processor, frame delay as a time shift between the received trajectory and the transmitted trajectory.   
     
     
         2 . The method of  claim 1 , wherein the reference features that are characteristic of content comprise faces. 
     
     
         3 . The method of  claim 1 , wherein the reduced reference features from the received video stream and the transmitted video stream comprise locations of the reference features. 
     
     
         4 . The method of  claim 1 , wherein the reduced reference features from the received video stream and the transmitted video stream comprise a location of rectangular areas surrounding the reference features. 
     
     
         5 . The method of  claim 1 , wherein the reduced reference features from the received video stream are calculated at less than a full frame rate of the received video stream. 
     
     
         6 . The method of  claim 1 , wherein the reduced reference features from the received video stream are calculated at no more than five frames per second. 
     
     
         7 . The method of  claim 1 , wherein the reduced reference features of the transmitted video stream are received via a communication channel that is separate from a communication channel used to transport the transmitted video stream. 
     
     
         8 . The method of  claim 1 , further comprising:
 receiving, by the processor, a frame of the transmitted video stream to the telecommunications endpoint, once per second, via a reliable channel; and   calculating, by the processor, reduced reference features from the frame.   
     
     
         9 . The method of  claim 1 , wherein the received video stream originates from a remotely-located camera accessible via a communication path. 
     
     
         10 . The method of  claim 1 , further comprising:
 detecting, by the processor, an absence of reference features from the received video stream.   
     
     
         11 . A system to calculate video frame delay in a video stream received by a telecommunications endpoint, the system comprising:
 a computer-readable storage medium, storing executable instructions; and   a processor coupled to the computer-readable storage medium, the processor, when executing the executable instructions:
 locates reference features that are characteristic of content in the received video stream; 
 calculates reduced reference features from the located reference features; 
 receives reduced reference features of a transmitted video stream, the transmitted video stream corresponding to the received video stream; 
 calculates a distance between the reduced reference features in the received video stream and the reduced reference features of the transmitted video stream; 
 calculates a received trajectory of the reduced reference features from the received video stream; 
 calculates a transmitted trajectory of the reduced reference features from the transmitted video stream; and 
 calculates video frame delay as a time shift between the received trajectory and the transmitted trajectory. 
   
     
     
         12 . The system of  claim 11 , wherein the reference features that are characteristic of content comprise faces. 
     
     
         13 . The system of  claim 11 , wherein the reduced reference features from the received video stream and the transmitted video stream comprise a location of rectangular areas surrounding the reference features. 
     
     
         14 . The system of  claim 11 , wherein the reduced reference features from the received video stream are calculated at less than a full frame rate of the received video stream. 
     
     
         15 . The system of  claim 11 , wherein the reduced reference features from the received video stream are calculated at no more than five frames per second. 
     
     
         16 . The system of  claim 11 , wherein the reduced reference features of the transmitted video stream are received via a communication channel that is separate from a communication channel used to transport the transmitted video stream. 
     
     
         17 . The system of  claim 11 , wherein the received video stream originates from a remotely-located camera accessible via a communication path. 
     
     
         18 . The system of  claim 11 , wherein the processor, when executing the executable instructions:
 detects an absence of reference features from the received video stream.