CPC G06T 5/50 (2013.01) [G06T 5/005 (2013.01); G06T 7/248 (2017.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/20221 (2013.01)] | 10 Claims |
1. A method of providing an image using a trained artificial intelligence (AI) model, the method comprising:
obtaining, using a camera of an electronic device, a plurality of images including a first image that includes a plurality of objects;
identifying a first object, which represents a first human, as a main object among the plurality of objects included in the first image;
detecting a movement of a second object, which represents a second human, based on a plurality of positions of the second object in at least two images of the plurality of images, the second object being different from the first object;
based on the detecting of the movement of the second object, determining, by a first trained AI model stored in a memory of the electronic device, the second object as an object for removal from the first image;
obtaining a second image including the first object by generating data for an area of the first image corresponding to the second object in order to remove the second object from the first image, wherein the generated data corresponds to a background hidden by the second object; and
displaying the obtained second image using a display of the electronic device.
|