An Effective Novel Neural Architecture for Synthesising Faces from given Text Description

Key Points

• The study aims to develop a novel neural architecture that can create realistic face images based on textual descriptions.
• Designing a new neural architecture
• Training on diverse datasets of text and facial images
• Evaluating performance using standard benchmarks
• Successfully generated high-quality face images from text inputs
• Demonstrated improved accuracy compared to existing models
• Achieved faster synthesis times, enhancing user experience