[StackGAN++] Realistic Image Synthesis with Stacked Generative Adversarial Networks | AISC
56Toronto Deep Learning Series, 15 October 2018
For slides and more information, visit https://tdls.a-i.science/events/2018-10-15/
Paper Review: https://arxiv.org/abs/1710.10916
Speaker: Michael O. Vertolli (Carleton University)
Host: Loblaw Digital
Date: Oct 15th, 2018
StackGAN++: Realistic Image Synthesis with Stacked Generative Adversarial Networks
Although Generative Adversarial Networks (GANs) have shown remarkable success in various tasks, they still face challenges in generating high quality images. In this paper, we propose Stacked Generative Adversarial Networks (StackGAN) aiming at generating high-resolution photo-realistic images. First, we propose a two-stage generative adversarial network architecture, StackGAN-v1, for text-to-image synthesis. The Stage-I GAN sketches the primitive shape and colors of the object based on given text description, yielding low-resolution images. The Stage-II GAN takes Stage-I results and text descriptions as inputs, and generates high-resolution images with photo-realistic details. Second, an advanced multi-stage generative adversarial network architecture, StackGAN-v2, is proposed for both conditional and unconditional generative tasks. Our StackGAN-v2 consists of multiple generators and discriminators in a tree-like structure; images at multiple scales corresponding to the same scene are generated from different branches of the tree. StackGAN-v2 shows more stable training behavior than StackGAN-v1 by jointly approximating multiple distributions. Extensive experiments demonstrate that the proposed stacked generative adversarial networks significantly outperform other state-of-the-art methods in generating photo-realistic images.