Photographic Text-to-Image Synthesis with a Hierarchically-Nested Adversarial Network

This paper presents a novel method to deal with the challenging task of generating photographic images conditioned on semantic image descriptions. Our method introduces accompanying hierarchical-nested adversarial objectives inside the network hierarchies, which regularize mid-level representations...

Full description

Saved in:
Bibliographic Details
Published in2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition pp. 6199 - 6208
Main Authors Zhang, Zizhao, Xie, Yuanpu, Yang, Lin
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.06.2018
Subjects
Gallium nitride
Generators
Image resolution
Measurement
Semantics
Task analysis
Training
Online AccessGet full text
ISSN1063-6919
DOI10.1109/CVPR.2018.00649

Cover

More Information
Summary:This paper presents a novel method to deal with the challenging task of generating photographic images conditioned on semantic image descriptions. Our method introduces accompanying hierarchical-nested adversarial objectives inside the network hierarchies, which regularize mid-level representations and assist generator training to capture the complex image statistics. We present an extensile single-stream generator architecture to better adapt the jointed discriminators and push generated images up to high resolutions. We adopt a multi-purpose adversarial loss to encourage more effective image and text information usage in order to improve the semantic consistency and image fidelity simultaneously. Furthermore, we introduce a new visual-semantic similarity measure to evaluate the semantic consistency of generated images. With extensive experimental validation on three public datasets, our method significantly improves previous state of the arts on all datasets over different evaluation metrics.
ISSN:1063-6919
DOI:10.1109/CVPR.2018.00649