Directional GAN: A Novel Conditioning Strategy for Generative Networks

Abstract

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.

arXiv commentAccepted to AICC workshop at CVPR 2021

References in corpus (9)

• Conditional Generative Adversarial Nets
• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
• Progressive Growing of GANs for Improved Quality, Stability, and Variation
• Adversarial Feature Learning
• Pose Guided Person Image Generation
• Precise Recovery of Latent Vectors from Generative Adversarial Networks
• Towards Multi-pose Guided Virtual Try-on Network
• Pose Guided Fashion Image Synthesis Using Deep Generative Model