Fader Networks: Generating Image Variations by Sliding Attribute Values

Abstract : This paper introduces a new encoder-decoder architecture that is trained to reconstruct images by disentangling the salient information of the image and the values of attributes directly in the latent space. As a result, after training, our model can generate different realistic versions of an input image by varying the attribute values. By using continuous attribute values, we can choose how much a specific attribute is perceivable in the generated image. This property could allow for applications where users can modify an image using sliding knobs, like faders on a mixing console, to change the facial expression of a portrait, or to update the color of some objects. Compared to the state-of-the-art which mostly relies on training adversarial networks in pixel space by altering attribute values at train time, our approach results in much simpler training schemes and nicely scales to multiple attributes. We present evidence that our model can significantly change the perceived value of the attributes while preserving the naturalness of images.
Document type :
Conference papers
Liste complète des métadonnées

Contributor : Lip6 Publications <>
Submitted on : Monday, December 10, 2018 - 10:53:11 AM
Last modification on : Thursday, March 21, 2019 - 1:12:03 PM


  • HAL Id : hal-01949501, version 1


Guillaume Lample, Neil Zeghidour, Nicolas Usunier, Antoine Bordes, Ludovic Denoyer, et al.. Fader Networks: Generating Image Variations by Sliding Attribute Values. 31st Conference on Neural Information Processing Systems (NIPS 2017), Dec 2017, Long Beach, CA, United States. pp.5969-5978. ⟨hal-01949501⟩



Record views