A Study on the Comparison and Measurement of Improved CartoonGAN Performance for High-Quality Image Generation

With the rapid growth of computer vision, among them, generative artificial intelligence is being used in various industries. CartoonGAN, one of the generating artificial intelligence, can create new images by changing the original image into a cartoon style, which can be used in various fields such as webtoons and animations. To this end, a process of improving the performance of CartoonGAN is necessary, and research for this should be conducted. In this study, to solve the performance degradation that occurs when CartoonGAN is applied, we would like to suggest a direction to improve by applying technologies such as SRCNN and Tiling.