| CPC H04N 23/632 (2023.01) [G06T 7/20 (2013.01); H04N 23/64 (2023.01)] | 20 Claims |

|
1. A method comprising:
receiving, by a processing device of a user device, a first preview image of a scene;
processing the first preview image using a first machine learning model to determine at least one direction to move the user device based on a composition score of the first preview image and at least one composition score of a plurality of candidate images, wherein each candidate image of the plurality of candidate images includes a first portion that comprises a subset of the first preview image and a second portion that comprises synthesized pixels;
detecting, using the processing device, movement of the user device; and
receiving, by the processing device, a second preview image of the scene after movement of the user device.
|