Self-Attention Generative Adversarial Networks

21 May 2018Han Zhang • Ian Goodfellow • Dimitris Metaxas • Augustus 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. In SAGAN, details can be generated using cues from all feature locations.

Full paper

Evaluation


Task Dataset Model Metric name Metric value Global rank Compare
Conditional Image Generation ImageNet 128x128 Self-attention FID 18.65 # 2
Conditional Image Generation ImageNet 128x128 Self-attention Inception score 52.52 # 2