CPC H04N 23/61 (2023.01) [G06N 20/00 (2019.01); G06T 7/74 (2017.01); G06T 7/80 (2017.01); G06T 2207/20081 (2013.01); G06T 2207/30244 (2013.01)] | 11 Claims |
1. A system comprising:
a non-transitory, computer readable storage medium having stored thereon a computer program comprising processor-implementable instructions; and a computer processor, which when executing the computer program, causes the system to implement:
a virtual camera operable to capture images of a scene;
an identification unit configured to identify an object of interest in images of the scene;
a pose processor configured to obtain a pose of the object of interest in the scene relative to the virtual camera;
a scene analyser operable to determine, based on at least one of the obtained pose of the object of interest and images captured by the virtual camera, a scene quality associated with images captured by the virtual camera at a respective pose; wherein the scene analyser comprises a first machine learning model trained to determine the scene quality associated with the images captured by the virtual camera at respective poses, where the first machine learning model is trained to recognize poses of high scene quality through analysis of images of high popularity in a user community; and
a controller configured to cause a pose of the virtual camera to be adjusted based on a determination that the scene quality of an image captured at a current pose is less than a threshold value, wherein:
the identification unit is configured to identify a type of scene that the images captured by the virtual camera corresponds to,
the scene analyser is further configured to determine a scene quality associated with the images captured by the virtual camera at a respective pose, based on the identified scene type,
the identification unit comprises a third machine learning model trained to identify a type of scene that the images captured by the virtual camera corresponds to, and
the third machine learning model is trained with images of different types of scene and corresponding scene identifiers.
|