Self-Attention Generative Adversarial Networks

05/21/2018
by   Han Zhang, et al.
0

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. Moreover, the discriminator can check that highly detailed features in distant portions of the image are consistent with each other. Furthermore, recent work has shown that generator conditioning affects GAN performance. Leveraging this insight, we apply spectral normalization to the GAN generator and find that this improves training dynamics. The proposed SAGAN achieves the state-of-the-art results, boosting the best published Inception score from 36.8 to 52.52 and reducing Frechet Inception distance from 27.62 to 18.65 on the challenging ImageNet dataset. Visualization of the attention layers shows that the generator leverages neighborhoods that correspond to object shapes rather than local regions of fixed shape.

READ FULL TEXT

page 2

page 6

page 7

page 8

research
09/26/2021

Logo Generation Using Regional Features: A Faster R-CNN Approach to Generative Adversarial Networks

In this paper we introduce Local Logo Generative Adversarial Network (LL...
research
06/14/2021

Improved Transformer for High-Resolution GANs

Attention-based models, exemplified by the Transformer, can effectively ...
research
04/08/2020

Attentive Normalization for Conditional Image Generation

Traditional convolution-based generative adversarial networks synthesize...
research
11/22/2018

TGANv2: Efficient Training of Large Models for Video Generation with Multiple Subsampling Layers

In this paper, we propose a novel method to efficiently train a Generati...
research
11/27/2019

Your Local GAN: Designing Two Dimensional Local Attention Mechanisms for Generative Models

We introduce a new local sparse attention layer that preserves two-dimen...
research
04/26/2021

CAGAN: Text-To-Image Generation with Combined Attention GANs

Generating images according to natural language descriptions is a challe...
research
04/12/2022

A Post Auto-regressive GAN Vocoder Focused on Spectrum Fracture

Generative adversarial networks (GANs) have been indicated their superio...

Please sign up or login with your details

Forgot password? Click here to reset