Publications
Detailed Information
Language-agnostic Semantic Consistent Text-to-Image Generation
Cited 1 time in
Web of Science
Cited 0 time in Scopus
- Authors
- Issue Date
- 2022-05
- Publisher
- ASSOC COMPUTATIONAL LINGUISTICS-ACL
- Citation
- PROCEEDINGS OF THE 1ST WORKSHOP ON MULTILINGUAL MULTIMODAL LEARNING (MML 2022), pp.1-5
- Abstract
- Recent GAN-based text-to-image generation models have advanced that they can generate photo-realistic images matching semantically with descriptions. However, research on multilingual text-to-image generation has not been carried out yet much. There are two problems when constructing a multilingual text-to-image generation model: 1) language imbalance issue in text-to-image paired datasets and 2) generating images that have the same meaning but are semantically inconsistent with each other in texts expressed in different languages. To this end, we propose a Language-agnostic Semantic Consistent Generative Adversarial Network (LaSC-GAN) for text-to-image generation, which can generate semantically consistent images via language-agnostic text encoder and Siamese mechanism. Experiments on relatively low resource language text image datasets show that the model has comparable generation quality as images generated by high-resource language text, and generates semantically consistent images for texts with the same meaning even in different languages.
- Files in This Item:
Item View & Download Count
Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.