Personalized content such as app backgrounds, profile pictures, etc. help in creating a rich and interactive user experience for an app, potentially driving an increase in the active usage. Recent advancements in generative imagery can be used to generate images that are personalized to the user. However, current image generation techniques require a user to explicitly provide prompts to the generation model. This disclosure describes techniques that use generative artificial intelligence to automatically generate customized images for use as an app background or in other contexts. Prompts for image creation are generated based on user-permitted contextual information by using a large language model or other technique. The generated images are filtered by an image recommendation model and provided for user selection to customize the app experience.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.