Text to Image Generation using GAN
This work presents a new retrieval system based on the generative adversarial network that will improve performance by easy processes on criminal face generation that extract facial traits from text descriptions and create a realistic human face.
Abstract
Text-to-image creation with generative adversarial networks (GAN) is a deep learning model that can produce images from text descriptions. For several years, due to low technological resources, police officials required a sketch artist to get the face of a criminal. So, to overcome this time-consuming process the proposed approach uses textual inputs of human facial traits and give the corresponding image of criminal as output. Artificial intelligence (AI) that links natural language processing and image processing have a challenge with converting information between text and image (NLP). As a result, we present a new retrieval system based on the generative adversarial network (GAN) that will improve performance by easy processes. Our work primarily focus on criminal face generation that extract facial traits from text descriptions and create a realistic human face.