MC-GAN: Multi-conditional Generative Adversarial Network for Image Synthesis

Abstract

In this paper, we introduce a new method for generating an object image from text attributes on a desired location, when the base image is given. One step further to the existing studies on text-to-image generation mainly focusing on the object's appearance, the proposed method aims to generate an object image preserving the given background information, which is the first attempt in this field. To tackle the problem, we propose a multi-conditional GAN (MC-GAN) which controls both the object and background information jointly. As a core component of MC-GAN, we propose a synthesis block which disentangles the object and background information in the training stage. This block enables MC-GAN to generate a realistic object image with the desired background by controlling the amount of the background information from the given base image using the foreground information from the text attributes. From the experiments with Caltech-200 bird and Oxford-102 flower datasets, we show that our model is able to generate photo-realistic images with a resolution of 128 x 128. The source code of MC-GAN is released.

BMVC 2018 accepted

References in corpus (1)

• Semantic Image Synthesis via Adversarial Learning

Cited by in corpus (8)

• Data synthesis and adversarial networks: A review and meta-analysis in cancer imaging
• A Survey and Taxonomy of Adversarial Neural Networks for Text-to-Image Synthesis
• Cascade Attention Guided Residue Learning GAN for Cross-Modal Translation
• A Layer-Based Sequential Framework for Scene Generation with GANs
• Words as Art Materials: Generating Paintings with Sequential GANs
• Pathology-Aware Generative Adversarial Networks for Medical Image Augmentation
• MeronymNet: A Hierarchical Approach for Unified and Controllable Multi-Category Object Generation
• OPAL-Net: A Generative Model for Part-based Object Layout Generation