Download pdf
A Comparative Analysis of StackGAN and AttnGAN in Text-to-Image Generation
- 1 Software Department, Shandong University, Shandong, China
* Author to whom correspondence should be addressed.
Abstract
This research looks at text-to-image generation as a whole, comparing two popular models—Stacked Generative Adversarial Networks (StackGAN) and Attentional Generative Adversarial Networks (AttnGAN)—and their respective strengths and weaknesses. Text-to-image generation has seen significant advancements with the introduction of GAN-based models, and this paper aims to explore how these models perform in terms of image quality, realism, and alignment with textual descriptions. Using the Caltech-UCSD Birds (CUB)-200-2011 dataset, which consists of bird images, extensive experiments were conducted to evaluate and compare the capabilities of the two models. The results indicate that AttnGAN outperforms StackGAN across multiple metrics, particularly in the accuracy of detail alignment and overall image realism. AttnGAN's multi-level attention mechanism allows it to pay attention to specific textual elements when generating related sections of the image, resulting in more aesthetically pleasing and semantically consistent outputs. Despite these advancements, challenges remain in improving both the diversity and quality of generated images. This work offers substantial insights into the capabilities and constraints of existing models, providing guidance for future research with the aim of improving text-to-image generation.
Keywords
Text-to-Image Generation, StackGAN, AttnGAN, Attention Mechanism.
[1]. Goodfellow I Pouget-Abadie J Mirza M et al. (2014). Generative adversarial net. Advances in neural information processing systems, 27
[2]. Reed S Akata Z Yan X et al. (2016). Generative adversarial text to image synthesis. International conference on machine learning, 1060-1069
[3]. Denton E L Chintala S Fergus R. (2015). Deep generative image models using laplacian pyramid of adversarial networks. Advances in neural information processing systems, 28
[4]. Radford A. (2015). Unsupervised representation learning with deep convolutional generative adversarial networks. arXiv preprint 1511.06434
[5]. Zhang H Xu T Li H et al. (2017). Stackgan: Text to photo-realistic image synthesis with stacked generative adversarial networks. Proceedings of the IEEE international conference on computer vision, 5907-5915
[6]. Reed S Akata Z Yan X et al. (2016). Generative adversarial text to image synthesis. International conference on machine learning, 1060-1069
[7]. Reed S E Akata Z Mohan S et al. (2016). Learning what and where to draw. Advances in neural information processing systems, 29
[8]. Fang H Gupta S Iandola F et al. (2015). From captions to visual concepts and back. Proceedings of the IEEE conference on computer vision and pattern recognition, 1473-1482
[9]. Xu K Ba J Kiros R et al. (2015). Show, attend and tell: Neural image caption generation with visual attention. International conference on machine learning, 2048-2057
[10]. Xu T Zhang P Huang Q et al. (2018). Attngan: Fine-grained text to image generation with attentional generative adversarial networks. Proceedings of the IEEE conference on computer vision and pattern recognition, 1316-1324
Cite this article
Wang,R. (2024). A Comparative Analysis of StackGAN and AttnGAN in Text-to-Image Generation. Applied and Computational Engineering,105,9-15.
Data availability
The datasets used and/or analyzed during the current study will be available from the authors upon reasonable request.
Disclaimer/Publisher's Note
The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of EWA Publishing and/or the editor(s). EWA Publishing and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
About volume
Volume title: Proceedings of CONF-MLA 2024 Workshop: Neural Computing and Applications
© 2024 by the author(s). Licensee EWA Publishing, Oxford, UK. This article is an open access article distributed under the terms and
conditions of the Creative Commons Attribution (CC BY) license. Authors who
publish this series agree to the following terms:
1. Authors retain copyright and grant the series right of first publication with the work simultaneously licensed under a Creative Commons
Attribution License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this
series.
2. Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the series's published
version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial
publication in this series.
3. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and
during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See
Open access policy for details).