Systems and methods for autodirecting a real-time transmission

Abstract

In some aspects, the described systems and methods provide for a system for processing a stream for real-time transmission. The system comprises a processor in communication with memory. The processor is configured to execute instructions for an autodirection component stored in memory that cause the processor to receive a real-time stream for an artistic performance, detect one or more human persons in the real-time stream, rank the detected one or more human persons in the real-time stream, select, based on the ranking, a subject from the detected one or more human persons, determine a subject framing for the real-time stream based on the selected subject, process the real-time stream to select a portion of each frame in the real-time stream according to the subject framing, wherein the portion of each frame includes at least the subject, and transmit the processed stream in real-time.

What is claimed is: 
     
         1 . A computer-implemented method for processing a real-time video stream, the method comprising, by a processor:
 receiving a first real-time video stream of an artistic performance from a first video source;   receiving a second real-time video stream of the artistic performance from a second video source different than the first video source;   time-synchronizing the first and second real-time video streams;   detecting one or more human persons in the first real-time video stream;   ranking the detected one or more human persons in the first real-time video stream;   selecting, based on the ranking, a subject from the detected one or more human persons;   determining a subject framing for the first real-time video stream based on the selected subject;   determining a subject framing for the second real-time video stream based on the selected subject;   processing the first real-time video stream and the second real-time video stream to select a portion of each frame in the real-time video stream and the second real-time video stream according to the subject framing, wherein the portion of each frame includes at least the subject;   and   generating an output video stream from the first real-time video stream and the second real-time video stream based on the selected portion of each frame.   
     
     
         2 . The method of  claim 1 , wherein the second video source is a smartphone camera. 
     
     
         3 . The method of  claim 1 , wherein the step of time-synchronizing the first and second video streams is performed based on audio signals in the first and second video streams. 
     
     
         4 . The method of  claim 1 , wherein the step of time-synchronizing the first and second real-time video streams synchronizes each of the first and second real-time video streams to a current time. 
     
     
         5 . The method of  claim 1 , wherein the detected one or more human persons are ranked based on proximity to the first and/or second video sources. 
     
     
         6 . The method of  claim 1 , wherein the detected one or more human persons are ranked based on a determination of which human person is speaking or singing in the artistic performance. 
     
     
         7 . The method of  claim 1 , wherein the step of generating the output video stream further comprises:
 after passage of a threshold time subsequent to an initial transmission of the output video stream, selecting a portion of the second real-time video stream for inclusion in the output video stream.   
     
     
         8 . The method of  claim 7 , further comprising:
 automatically selecting, by a computer processor, portions of the first real-time video stream and the second real-time video stream to include in the output video stream.   
     
     
         9 . The method of  claim 8 , further comprising, for at least one selected portion of the first real-time video stream or the second real-time video stream, adjusting a zoom of the at least one selected portion prior to including the at least one selected portion in the output video stream. 
     
     
         10 . The method of  claim 8 , wherein the automatic selection is based on a determination that the subject is no longer present in the selected framing of the first real-time video stream or in the selected framing of the second real-time video stream. 
     
     
         11 . A system for generating a video stream for real-time transmission, the system comprising a processor in communication with memory, the processor being configured to execute instructions for an autodirection component stored in memory that cause the processor to:
 receive a first real-time video stream of an artistic performance from a first video source;   receive a second real-time video stream of the artistic performance from a second video source different than the first video source;   time-synchronize the first and second real-time video streams;   detect one or more human persons in the first real-time video stream;   rank the detected one or more human persons in the first real-time video stream;   select, based on the ranking, a subject from the detected one or more human persons;   determine a subject framing for the first real-time video stream based on the selected subject;   determine a subject framing for the second real-time video stream based on the selected subject;   process the first real-time video stream and the second real-time video stream to select a portion of each frame in the real-time video stream and the second real-time video stream according to the subject framing, wherein the portion of each frame includes at least the subject;   and   generate an output video stream from the first real-time video stream and the second real-time video stream based on the selected portion of each frame.   
     
     
         12 . The system of  claim 11 , wherein the second video source is a smartphone camera. 
     
     
         13 . The system of  claim 11 , wherein the step of time-synchronizing the first and second video streams is performed based on audio signals in the first and second video streams. 
     
     
         14 . The system of  claim 11 , wherein the step of time-synchronizing the first and second real-time video streams synchronizes each of the first and second real-time video streams to a current time. 
     
     
         15 . The system of  claim 11 , wherein the detected one or more human persons are ranked based on proximity to the first and/or second video sources. 
     
     
         16 . The system of  claim 11 , wherein the detected one or more human persons are ranked based on a determination of which human person is speaking or singing in the artistic performance. 
     
     
         17 . The system of  claim 11 , wherein the step of generating the output video stream further comprises:
 after passage of a threshold time subsequent to an initial transmission of the output video stream, selecting a portion of the second real-time video stream for inclusion in the output video stream.   
     
     
         18 . The system of  claim 17 , the processor further configured to:
 automatically select portions of the first real-time video stream and the second real-time video stream to include in the output video stream.   
     
     
         19 . The system of  claim 18 , the processor further configured to, for at least one selected portion of the first real-time video stream or the second real-time video stream, adjust a zoom of the at least one selected portion prior to including the at least one selected portion in the output video stream. 
     
     
         20 . The system of  claim 18 , wherein the automatic selection is based on a determination that the subject is no longer present in the selected framing of the first real-time video stream or in the selected framing of the second real-time video stream.