Deep Automodulators
This provides a useful building block for image applications and other data domains by enabling new capabilities like style-mixing with real input images.
The authors introduced automodulators, a new category of generative autoencoders that can reproduce individual real-world images and generate fused samples from arbitrary combinations of images, enabling instantaneous style-mixing. They demonstrated state-of-the-art results in autoencoder comparison and visual image quality nearly indistinguishable from state-of-the-art GANs on four image datasets.
We introduce a new category of generative autoencoders called automodulators. These networks can faithfully reproduce individual real-world input images like regular autoencoders, but also generate a fused sample from an arbitrary combination of several such images, allowing instantaneous 'style-mixing' and other new applications. An automodulator decouples the data flow of decoder operations from statistical properties thereof and uses the latent vector to modulate the former by the latter, with a principled approach for mutual disentanglement of decoder layers. Prior work has explored similar decoder architecture with GANs, but their focus has been on random sampling. A corresponding autoencoder could operate on real input images. For the first time, we show how to train such a general-purpose model with sharp outputs in high resolution, using novel training techniques, demonstrated on four image data sets. Besides style-mixing, we show state-of-the-art results in autoencoder comparison, and visual image quality nearly indistinguishable from state-of-the-art GANs. We expect the automodulator variants to become a useful building block for image applications and other data domains.