On Conditioning GANs to Hierarchical Ontologies
This work addresses the problem of enhancing image synthesis fidelity for fashion applications, representing an incremental improvement in domain-specific GAN conditioning.
The paper tackles the challenge of generating high-quality fashion images from text descriptions by proposing Ontology Generative Adversarial Networks (O-GANs) that condition on a hierarchical fashion ontology, resulting in improved image quality as measured by Fréchet Inception Distance and Inception Score, and better conditioning results evaluated by implicit similarity between text and generated images.
The recent success of Generative Adversarial Networks (GAN) is a result of their ability to generate high quality images from a latent vector space. An important application is the generation of images from a text description, where the text description is encoded and further used in the conditioning of the generated image. Thus the generative network has to additionally learn a mapping from the text latent vector space to a highly complex and multi-modal image data distribution, which makes the training of such models challenging. To handle the complexities of fashion image and meta data, we propose Ontology Generative Adversarial Networks (O-GANs) for fashion image synthesis that is conditioned on an hierarchical fashion ontology in order to improve the image generation fidelity. We show that the incorporation of the ontology leads to better image quality as measured by Fréchet Inception Distance and Inception Score. Additionally, we show that the O-GAN achieves better conditioning results evaluated by implicit similarity between the text and the generated image.