9th International Conference on ICT for Sustainable Development

Authors - Sharmila Sengupta, Gauri Nagral, Meera Sawantdesai, Kritika Yadav, Yashraj Mulwani
Abstract - In this day and age, advertisements play a huge role for businesses to flourish in competitive economies. It is said that an attractive ad is worth more than a thousand words. Creating an advertisement involves conceptualizing an idea, designing visual and textual content, and strategically placing it across media platforms to reach and influence the target audience. This entire process can take anywhere from a few hours to several months, depending on the complexity and medium of the advertisement. However, the tremendous growth in technology can turn those hours and months into just a few seconds. This paper explores the capabilities of Generative Artificial Intelligence (GenAI) for visual element generation through advanced techniques like text-to-image generation. It analyzes leading models like StackGAN, Stable Diffusion - DreamBooth, and LoRA, identifying the most efficient among them. The research then focuses on implementing these models to automate advertisement creation for product-based companies, aiming to replace manual efforts in the creative department while maintaining privacy and security of proprietary company information. The proposed system, ADGenAI offers insights into the transformative potential of GenAI in streamlining creative processes, fostering efficiency, and envisioning the future of the advertising industry. Specifically designed to alleviate the pressures faced by creative teams, such as the demand for high-quality content under tight deadlines, this system enhances creativity and productivity by automating the initial stages of ad production and enabling the exploration of new advertising concepts with minimal resource investment.