SCENE-BASED TEXT-TO-IMAGE GENERATION WITH HUMAN PRIORS

• Main Authors: GAFNI, Oran, POLYAK, Adam, TAIGMAN, Yaniv Nechemia
• Format: Patent
• Language: English; French
• Published: 11.07.2024
• Subjects: CALCULATING; COMPUTING; COUNTING; PHYSICS

In one embodiment, a method includes accessing a text input and a scene input corresponding to the text input, wherein the scene input comprises semantic segmentations, generating text tokens for the text input and scene tokens for the scene input by machine-learning models, generating predicted ima...