Cyclic-GAN and its application

Montag, 28. Juni 2021 44

Logo des Projekts CyclicGan

For this topic we introduced a new idea about 3D medical images segmentation and super-resolution. Because of the large size of high-resolution medical images will cost a lot of time to train and calculate
So we hope to use low-resolution images to create high-resolution images using a deep learning model. And also purposed a new way to do 3D medical images segmentation. In our method we used Cyclic-GAN which is purposed in ICCV 2017 by Zhu Junyan et al. to finish our tasks.


  • Haoze Jing