Animating Text Descriptions into Characters: A Comparative Review of Generative Models | ||
| IJCI. International Journal of Computers and Information | ||
| Article 4, Volume 12, Issue 1, January 2025, Pages 43-66 PDF (2.12 M) | ||
| Document Type: Original Article | ||
| DOI: 10.21608/ijci.2024.307030.1167 | ||
| Authors | ||
| Sameh Zarif* 1; Khalid M. Amin2; Abdalfatah Najjar3; Marian Wagdy4 | ||
| 1Faculty of computers and information, Menofia university | ||
| 2Information technology dept., Faculty of computers and information, Menofia university | ||
| 3Information Technology Department | ||
| 4Department of Information Technology, Faculty of Computers and Information | ||
| Abstract | ||
| In recent times, the advent of text-to-image generative AI technologies, commonly referred to as AI Image Generators, has captured widespread attention for their remarkable capability to swiftly produce visuals based on textual descriptions. This development has ignited diverse perspectives, especially within the animation sector, making it a focal point for scholarly investigation due to the introduction of generative adversarial networks. Despite the progress, the domain confronts several challenges that necessitate further scholarly inquiry, including the production of high-resolution imagery featuring multiple elements and the creation of evaluation metrics that align with human assessment. Moreover, the outcomes of this study reveal that AI Image Generators hold the potential to substantially boost creative outputs in animation by providing a variety of characters and settings with superior visual quality. This review aims to examine and compare the extensive body of work in this field comprehensively. It will detail the algorithms employed, identify existing issues, and propose new directions for research. | ||
| Keywords | ||
| generative-art; AnimeGAN; text-to-image; animation-character; generative-adversarial-network | ||
|
Statistics Article View: 175 PDF Download: 135 |
||
