Auto-reframing and multi-cam functions of video editing application

A system comprising:

• one or more processors;
• and
• a non-transitory computer readable medium comprising one or more sequences of instructions which, when executed by the one or more processors, cause the one or more processors to perform operations comprising:
• obtaining a plurality of training data sets, wherein each particular training data set in the plurality of training data sets comprises:
• an original video;
• and
• a modified video, wherein a first plurality of frames from the original video are modified to generate corresponding frames in the modified video;
• training a machine learning model, using the plurality of training data sets, to generate modified videos from original videos such that one or more frames in the original videos are modified to generate corresponding frames in respective modified videos;
• obtaining a target original video;
• and
• applying the trained machine learning model to the target original video to generate a target modified video, wherein applying the trained machine learning model to the target original video comprises:
• identifying a second plurality of frames of the target original video;
• modifying each of the second plurality of frames to generate a modified second plurality of frames;
• and
• generating the target modified video based on the modified second plurality of frames.

A method comprising:

• obtaining a plurality of training data sets, wherein each particular training data set in the plurality of training data sets comprises:
• an original video;
• and
• a modified video, wherein a first plurality of frames from the original video are modified to generate corresponding frames in the modified video;
• training a machine learning model, using the plurality of training data sets, to generate modified videos from original videos such that one or more frames in the original videos are modified to generate corresponding frames in respective modified videos;
• obtaining a target original video;
• and
• applying the trained machine learning model to the target original video to generate a target modified video, wherein applying the trained machine learning model to the target original video comprises:
• identifying a second plurality of frames of the target original video;
• modifying each of the second plurality of frames to generate a modified second plurality of frames;
• and
• generating the target modified video based on the modified second plurality of frames.