| CPC G06T 15/00 (2013.01) [G06T 3/4046 (2013.01); G06T 2210/41 (2013.01)] | 10 Claims |

|
1. A learning device of an image generation model that, in a case in which at least one target image for a subject, which includes a specific structure, having at least one representation format, and target information representing a target representation format of the target image are input, derives a virtual image having the target representation format from the target image, the learning device comprising at least one processor,
wherein the image generation model includes
a first network that outputs a subject model representing the subject by deriving each feature amount of the target image having the at least one representation format and combining the feature amounts by inputting the target image,
a second network that, in a case in which the target information and the subject model are input, outputs a latent variable obtained by dimensionally compressing a feature of the subject model according to the target information, and
a third network that, in a case in which the target information, the subject model, and the latent variable are input, outputs the virtual image, and
wherein the processor is configured to train the first network, the second network, and the third network based on a plurality of teacher images having different representation formats for the subject including the specific structure, and a plurality of teacher data including specific teacher information representing a specific representation format among representation formats of the plurality of teacher images.
|