MISO: Mutual Information Loss with Stochastic Style Representations for Multimodal Image-to-Image Translation
arXiv:1902.03938
Abstract
Unpaired multimodal image-to-image translation is a task of translating a given image in a source domain into diverse images in the target domain, overcoming the limitation of one-to-one mapping. Existing multimodal translation models are mainly based on the disentangled representations with an image reconstruction loss. We propose two approaches to improve multimodal translation quality. First, we use a content representation from the source domain conditioned on a style representation from the target domain. Second, rather than using a typical image reconstruction loss, we design MILO (Mutual Information LOss), a new stochastically-defined loss function based on information theory. This loss function directly reflects the interpretation of latent variables as a random variable. We show that our proposed model Mutual Information with StOchastic Style Representation(MISO) achieves state-of-the-art performance through extensive experiments on various real-world datasets.
References in corpus (10)
- Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
- Large Scale GAN Training for High Fidelity Natural Image Synthesis
- Spectral Normalization for Generative Adversarial Networks
- Progressive Growing of GANs for Improved Quality, Stability, and Variation
- InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets
- The Unreasonable Effectiveness of Deep Features as a Perceptual Metric
- Autoencoding beyond pixels using a learned similarity metric
- Augmented CycleGAN: Learning Many-to-Many Mappings from Unpaired Data
- DualGAN: Unsupervised Dual Learning for Image-to-Image Translation
- Multimodal Unsupervised Image-to-Image Translation