DNA-GAN: Learning Disentangled Representations from Multi-Attribute Images
This addresses the challenge of learning disentangled representations for multi-attribute images, which is incremental as it builds on existing GAN-based methods to improve factors like identity preservation and image quality.
The paper tackles the problem of disentangling factors of variation in images by proposing DNA-GAN, a supervised learning model that learns DNA-like latent representations where each piece corresponds to an independent attribute, and it demonstrates effectiveness on Multi-PIE and CelebA datasets.
Disentangling factors of variation has become a very challenging problem on representation learning. Existing algorithms suffer from many limitations, such as unpredictable disentangling factors, poor quality of generated images from encodings, lack of identity information, etc. In this paper, we propose a supervised learning model called DNA-GAN which tries to disentangle different factors or attributes of images. The latent representations of images are DNA-like, in which each individual piece (of the encoding) represents an independent factor of the variation. By annihilating the recessive piece and swapping a certain piece of one latent representation with that of the other one, we obtain two different representations which could be decoded into two kinds of images with the existence of the corresponding attribute being changed. In order to obtain realistic images and also disentangled representations, we further introduce the discriminator for adversarial training. Experiments on Multi-PIE and CelebA datasets finally demonstrate that our proposed method is effective for factors disentangling and even overcome certain limitations of the existing methods.