Text-driven photo style adjustment with generative ai

Abstract

Systems and methods for text-driven photo style adjustment with generative artificial intelligence are provided. An example method includes receiving an image and text, determining and filtering one or more face bounding boxes and one or more body bounding boxes to obtain and analyze a further face bounding box and a further human body bounding box, generating text description based on the analysis, combining the text description and the text to obtain updated text, encoding the image into an image latent, encoding the updated text into a text embedding, generating a first noise, adding the first noise to the image latent to obtain a noisy image latent, providing the noisy image latent and the text embedding to a first neural network to generate a second noise, removing the second noise from the noisy image latent to obtain a denoised image latent, and decoding the denoised image latent into a result image.