Directional GAN: A Novel Conditioning Strategy for Generative Networks
This simplifies image content creation for marketers and designers by enabling conditional generation with existing unconditional GANs, though it is incremental as it builds on standard GAN frameworks.
The paper tackled the problem of generating images conditioned on semantic attributes without retraining the generator, proposing a novel conditioning strategy that modifies latent vectors using directional vectors in latent space, achieving an average accuracy of 86.4% across attributes on multiple datasets.
Image content is a predominant factor in marketing campaigns, websites and banners. Today, marketers and designers spend considerable time and money in generating such professional quality content. We take a step towards simplifying this process using Generative Adversarial Networks (GANs). We propose a simple and novel conditioning strategy which allows generation of images conditioned on given semantic attributes using a generator trained for an unconditional image generation task. Our approach is based on modifying latent vectors, using directional vectors of relevant semantic attributes in latent space. Our method is designed to work with both discrete (binary and multi-class) and continuous image attributes. We show the applicability of our proposed approach, named Directional GAN, on multiple public datasets, with an average accuracy of 86.4% across different attributes.