Giters

phillipi / pix2pix

Image-to-image translation with conditional adversarial nets

Home Page:https://phillipi.github.io/pix2pix/

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Input to generator

levindabhi opened this issue · comments

commented

Hey guys, So I am having segmentation map with 8 classes and I am inputting that with one RGB image to Pix2Pix generator. The output of a generator is an RGB image. What would be a good option to have

  1. Covert segm map into single-channel image and concat with RGB image and then input
  2. Make 8 channel segm tensor and concat with RGB image and then input