MISSING IMAGE DATA IMPUTATION METHOD USING NEURAL NETWORK AND APPARATUS THEREFOR

摘要

Disclosed are a missing image data imputation method using a neural network and an apparatus therefor. The missing image data imputation method according to an embodiment of the present invention comprises the steps of: receiving input image data for at least two domains among preset multiple domains; and reconstructing missing image data of a preset target domain by using a neural network that receives the input image data for at least two domains, wherein the neural network may be trained by combining real image data with fake image data of a first target domain, which is generated using input real image data for at least two of the multiple domains, and by using a multi-cycle consistency loss in which the real image data and an image reconstructed using the input combined image data must be similar to each other.