AttnGAN: Fine-Grained Text to Image Generation with Attentional Generative Adversarial Networks
TL
TL;DR
An Attentional Generative Adversarial Network that allows attention-driven, multi-stage refinement for fine-grained text-to-image generation and for the first time shows that the layered attentional GAN is able to automatically select the condition at the word level for generating different parts of the image.
Abstract
In this paper, we propose an Attentional Generative Adversarial Network (AttnGAN) that allows attention-driven, multi-stage refinement for fine-grained text-to-image generation. With a novel attentional generative network, the AttnGAN can synthesize fine-grained details at different sub-regions of the image by paying attentions to the relevant words in the natural language description. In addition, a deep attentional multimodal similarity model is proposed to compute a fine-grained image-text matching loss for training the generator. The proposed AttnGAN significantly outperforms the previous state of the art, boosting the best reported inception score by 14.14% on the CUB dataset and 170.25% on the more challenging COCO dataset. A detailed analysis is also performed by visualizing the attention layers of the AttnGAN. It for the first time shows that the layered attentional GAN is able to automatically select the condition at the word level for generating different parts of the image.
References41 items
1
Context Encoding for Semantic Segmentation
2
Densely Connected Pyramid Dehazing Network
3
Translating and Segmenting Multimodal Medical Volumes with Cycle- and Shape-Consistency Generative Adversarial Network
4
Photographic Text-to-Image Synthesis with a Hierarchically-Nested Adversarial Network
5