CPC G06V 20/13 (2022.01) [B64C 39/024 (2013.01); B64D 47/08 (2013.01); G01S 13/76 (2013.01); G01S 13/878 (2013.01); G06T 7/248 (2017.01); G06V 20/17 (2022.01); H04N 23/695 (2023.01); B64U 2101/30 (2023.01); B64U 2201/104 (2023.01)] | 20 Claims |
1. A method comprising:
capturing successive video image frames to generate a video;
processing the video image frames from an imaging device to determine previous positions of multiple subjects;
obtaining motion information of the multiple subjects from one or more sensors physically attached to the multiple subject respectively;
determining a predicted position of the multiple subjects respectively from the motion information of the multiple subjects, and the previous positions of the multiple subjects;
adjusting the imaging device based on the predicted position of the multiple subjects respectively so that the multiple subjects are subject is framed within the successive video image frames;
determining a region of interest of a subsequent image of the multiple subjects in a predefined spatial relationship relative to the predicted position;
obtaining subsequent successive video image frames with the imaging device;
processing the successive video image frames to determine an updated position of the multiple subjects so that the multiple subjects are automatically tracked by the imaging device;
monitoring a voice recognition system and interpreting audio or voice commands;
adjusting the trajectory based upon the audio or the voice commands;
identifying the multiple subjects within the successive video image frames; and
providing live videos being captured to a sharing service via an application of the imaging device.
|