Prompt engineering and generative AI for goal-based imagery
AssigneeThe Toronto-Dominion Bank
InventorBreena Patricia Gormley
Life of this patent
Estimating the term…
Abstract
An example operation may include one or more of storing a generative artificial intelligence (GenAI) model configured to create images, displaying a plurality of prompts on a user interface of a software application, receiving an identifier of a goal of the user and attributes of the goal via the plurality of prompts on the user interface, executing the GenAI model on the identifier of the goal of the user and the attributes of the goal to generate a custom image of the goal for the user, and displaying the custom image of the goal via the user interface of the software application.
Claims20
A computing system comprising:
- a memory;
- and
- a processor coupled to the memory, the processor configured to:
- display a plurality of prompts on a user interface of a software application via a display device on the computing system,
- receive an identifier of a goal associated with an object and a data structure associated with the goal via the user interface,
- execute a generative artificial intelligence (GenAI) model on the identifier of the goal to generate a custom image of the object formed from a collage of integrated images of the object,
- detect an amount of progress towards the goal based on changes to data within the data structure;
- and
- fill-in a corresponding subset of pixels to reveal a corresponding image segment of the custom image via the user interface based on execution of the GenAI model on the detected amount of progress.
A method comprising:
- displaying a plurality of prompts on a user interface of a software application via a display device of a computing system;
- receiving an identifier of a goal associated with an object and a data structure associated with the goal via the user interface;
- executing a generative artificial intelligence (GenAI) model on the identifier of the goal to generate a custom image of the object formed from a collage of integrated images of the object;
- detecting an amount of progress towards the goal based on changes to data within the data structure;
- and
- filling-in a corresponding subset of pixels to reveal a corresponding image segment of the custom image via the user interface based on execution of the GenAI model on the detected amount of progress.
A computer-readable medium comprising instructions stored therein which when executed by a processor cause the processor to perform:
- displaying a plurality of prompts on a user interface of a software application via a display device of a computing system;
- receiving an identifier of a goal of associated with an object and a data structure associated with the goal via the user interface;
- executing a generative artificial intelligence (GenAI) model on the identifier of the goal to generate a custom image of the formed from a collage of integrated images of the object;
- detecting an amount of progress towards the goal based on changes to data within the data structure;
- and
- filling-in a corresponding subset of pixels to reveal a corresponding image segment of the custom image via the user interface based on execution of the GenAI model on the detected amount of progress.
Dependencies are inferred from the claim text.