Official Tensorflow implementation of U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation (ICLR 2020)
I am not sure if someone asks this question before. If I trained A->B, can I used the model to generate B->A? Are there ways for me to convert the model, or do I have to train again to get B->A model? Thanks!
Hi guys,
I am not sure if someone asks this question before. If I trained A->B, can I used the model to generate B->A? Are there ways for me to convert the model, or do I have to train again to get B->A model? Thanks!