OPTIMIZATION FOR DNN CONPOSITION WITH REAL-TIME INFERENCE IN MOBILE ENVIRONMENT

摘要

Provided is a system for providing a deep neural network (DNN) model capable of real-time interference in a mobile environment. The comprises at least one processor implemented to execute a computer readable command, wherein the at least one processor includes a learning part learning a DNN-based style transfer model by using a specific style of image to be learned, and the style transfer model is a DNN model having a structure in which the number of deep layers is reduced through a transfer learning method using pre-learned result.