Target scene composition using generative ai

A system comprising:

• a memory component;
• and
• a processing device coupled to the memory component, the processing device to perform operations comprising:
• receiving a natural language description of an image to be generated using a machine learning model;
• extracting, from a first portion of the natural language description of the image to be generated, a control element from a plurality of control elements, wherein the control element is associated with a set of parameters, and wherein each control element of the plurality of control elements is associated with a different set of parameters;
• extracting, from a second portion of the natural language description of the image to be generated, a sub-prompt, wherein the sub-prompt defines a visual element of visual elements of the image to be generated, and wherein the second portion of the natural language description comprises natural language text different from the natural language text of the first portion of the natural language description of the image;
• identifying a relationship between the control element and the sub-prompt based on the natural language description of the image to be generated;
• and
• generating, by the machine learning model, an image based on the control element, the sub-prompt, and the relationship, wherein the image includes visual elements corresponding to the control element and the sub-prompt.

A non-transitory computer-readable medium storing executable instructions, which when executed by a processing device, cause the processing device to perform operations comprising:

• receiving a natural language description of an image and a source image;
• extracting, from a first portion of the natural language description of the image, a control element from a plurality of control elements, wherein the control element is associated with a set of parameters, and wherein each control element of the plurality of control elements is associated with a different set of parameters;
• extracting, from a second portion of the natural language description of the image, a sub-prompt, wherein the sub-prompt defines a visual element, and wherein the second portion of the natural language description comprises natural language text different from the natural language text of the first portion of the natural language description of the image;
• generating the visual element using a generative model, wherein the visual element is based on the sub-prompt;
• and
• arranging an image using the visual element and the source image.