PT Unknown
AU Yaxing Wang
   Hector Laria Mantecon
   Joost Van de Weijer
   Laura Lopez-Fuentes
   Bogdan Raducanu
TI TransferI2I: Transfer Learning for Image-to-Image Translation from Small Datasets
BT 19th IEEE International Conference on Computer Vision
PY 2021
BP 13990
EP 13999
DI 10.1109/ICCV48922.2021.01375
AB Image-to-image (I2I) translation has matured in recent years and is able to generate high-quality realistic images. However, despite current success, it still faces important challenges when applied to small domains. Existing methods use transfer learning for I2I translation, but they still require the learning of millions of parameters from scratch. This drawback severely limits its application on small domains. In this paper, we propose a new transfer learning for I2I translation (TransferI2I). We decouple our learning process into the image generation step and the I2I translation step. In the first step we propose two novel techniques: source-target initialization and self-initialization of the adaptor layer. The former finetunes the pretrained generative model (e.g., StyleGAN) on source and target data. The latter allows to initialize all non-pretrained network parameters without the need of any data. These techniques provide a better initialization for the I2I translation step. In addition, we introduce an auxiliary GAN that further facilitates the training of deep I2I systems even from small datasets. In extensive experiments on three datasets, (Animal faces, Birds, and Foods), we show that we outperform existing methods and that mFID improves on several datasets with over 25 points.
ER