To search, Click below search items. |
| All Published Papers Search Service |
|
|
|
Title |
Generating an Image from Arabic Text using Generative Network of Fine-Grained Visual Descriptions |
|
Author |
Sara Maher, Mohamed Loey |
| Citation |
Vol. 25 No. 1 pp. 195-201 |
|
Abstract |
Converting text representations in natural language into pictures is a complex computer vision issue and has many practical applications. Text-image does not vary from issues with language translation. In the same way similar semantics can be encoded in two different languages, images and text are two different languages to encode related information. None the less, these problems are totally different because text-image or image-text conversions are highly multimodal problems. In this paper, we propose our model for Arabic text description that allows multi-stage, attention-driven for refinement for fine-grained Arabic text-to-image generation. With a modern attentional generative network, The Attentional model enables fine-grained information to be synthesized in different image sub-regions by paying attention to the relevant terms in the definition of the natural Arabic language. We train the model from scratch to Modified-Arabic dataset. A word level fine-grained image-text matching loss computed by our Proposed-Method is the significant term in our Network. Two key neural networks that map sub-regions of the picture and Arabic words of the sentence to a common semantic space are learned by our Proposed-Method. On the Arabic text encoder and image encoder, our model achieves good efficiency, characterized by ease and accuracy in describing the images on the Caltech-UCSD Birds 200-2011 dataset. |
|
Keywords |
Machine Learning, Deep Learning, Generative Adversarial Networks, Recurrent Neural Network, Natural Language Processing, Text Analysis. |
|
URL |
