02-04-2020 дата публикации
Номер: US20200101375A1
Принадлежит:
A method of deep learning from real world and digital exemplars includes determining, by one or more processors of a computer system, a style component of a digital environment of a game platform, combining, by the one or more processors of the computer system, the style component with content derived from a real world exemplar, morphing, by one or more processors of a computer system, the real world exemplar to an augmented digital exemplar of the game platform, and adapting, by the one or more processors of the computer system, at least one deep learning algorithm to accomplish at least one of the determining, combining and morphing. 1. A method , the method comprising:determining, by one or more processors of a computer system, a style component of a digital environment of a game platform;combining, by the one or more processors of the computer system, the style component with content derived from a real world exemplar;morphing, by one or more processors of a computer system, the real world exemplar to an augmented digital exemplar of the game platform; andadapting, by the one or more processors of the computer system, at least one deep learning algorithm to accomplish at least one of the determining, combining and morphing.2. The method of claim 1 , further comprising:generating, by the one or more processors of the computer system, a comparison score for the morphed real world exemplar compared with the real world exemplar and at least one pure digital exemplar;using, by the one or more processors of the computer system, the comparison score as a training boosted rate; andoptimizing, by the one or more processors of the computer system, back propagation with the training boosted rate.3. The method of claim 1 , further comprising:training, by the one or more processors of the computer system,the at least one deep learning algorithm in a burst mode in response to a user changing from the digital environment to a second digital environment within the game platform ...
Подробнее