CycleGAN icon indicating copy to clipboard operation
CycleGAN copied to clipboard

Published 20 hours ago • verified publisher icon junyanz

The number of training dataset

Open qianxiao111 opened this issue 5 years ago • 3 comments

Hello, I have a question. How many images in the training dataset can be used to train an effective model? For example, domain X has 6000 undistorted images, domain Y has only 145 images with five distortion types. The scenes of Y are included in X . In this case, could generate images with five distortion types, which is similar to Y domain?

qianxiao111 avatar Mar 01 '19 13:03 qianxiao111

I will recommend 1000+ images per dataset. But you may want to try your dataset and see how it goes.

junyanz avatar Mar 25 '19 03:03 junyanz

Thanks, I have a question, if the Y domain constain five distortion types with one distortion level, is the generated images mixed distortions with five distortion types or single distortion?

qianxiao111 avatar Mar 25 '19 03:03 qianxiao111

probably multiple distortions.

junyanz avatar Mar 25 '19 04:03 junyanz