Self-Attention Generative Adversarial Networks

arXiv 2018 Han ZhangIan GoodfellowDimitris MetaxasAugustus Odena

In this paper, we propose the Self-Attention Generative Adversarial Network (SAGAN) which allows attention-driven, long-range dependency modeling for image generation tasks. Traditional convolutional GANs generate high-resolution details as a function of only spatially local points in lower-resolution feature maps... (read more)

PDF Abstract

Evaluation results from the paper


Task Dataset Model Metric name Metric value Global rank Compare
Conditional Image Generation ImageNet 128x128 Self-attention FID 18.65 # 4
Conditional Image Generation ImageNet 128x128 Self-attention Inception score 52.52 # 4