What algorithm to use

Question

I am stuck on what algorithm to use. I want to train my program on a dataset where i have an input image, and an output image which is a modified version of the input. The whole context of the image is important, and each pixel counts.

After training i want to input an image, and get an output similar to the training data.

Maybe you can try autoencoders? – TitoOrt Mar 12 '19 at 13:45 — TitoOrt, Mar 12 '19 at 13:45

score 0 · Answer 1 · answered Mar 12 '19 at 14:28

0

You can use Deep convolution GAN network of Deep learning. It would be a generative model.

you can get detailed help from :

https://medium.com/@keisukeumezawa/dcgan-generate-the-images-with-deep-convolutinal-gan-55edf947c34b

answered Mar 12 '19 at 14:28

Reeves

101
2

What algorithm to use

1 Answers1