Abstract
With user permission, advertisements that are delivered to a user can be selected based on user data in a manner that the advertisements are likely to be useful to the user. However, the content of the advertisement itself is not personalized to the viewer or viewer group/ market segment. This disclosure describes techniques that utilize generative artificial intelligence (gen AI) to automatically create video ad content customized to particular market segments. Inputs from the advertiser such as brand images, logos, props, etc., together with market segment descriptions are provided to a large language model (LLM) that generates text prompts. The generated text prompts are fed to a generative video model to obtain video ads tailored to the market segment targeted in the text input. The techniques enable automated creation of video ads tailored to multiple market segments without incurring high production costs. Video ads are made more relevant to the user, enabling the advertiser to attract and retain customers.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Muthamizh Selvan, Manoj, "Generative Video Ads Tailored to Market Segments", Technical Disclosure Commons, (April 10, 2024)
https://www.tdcommons.org/dpubs_series/6867