XGAN: UNSUPERVISED IMAGE-TO-IMAG
2018-01-26 本文已影响113人
klory
https://arxiv.org/pdf/1711.05139.pdf
Novelty
- Unsupervise
- Two domain share the same latent space
- predict class at encoder
Problem
- Only two domains at a time, need to train new model for new domain
Structure
Structure.pngLoss
Loss.pngDataset
Self-created CartoonSet dataset