Image data synthesis method, device, and product

Abstract

The application discloses an image data synthesis method, device and product, relates to the technical field of image processing, and comprises the following steps: acquiring original artistic text data from a data source; generating an image description subtitle by a large language model according to the original text data, and generating an image download task of the image description subtitle by an image generation model; executing the image download task to obtain a target image, and performing synthesis processing on the target image and metadata corresponding to the target image to obtain artistic text image data with file description. Therefore, the problem that the text and pixel modeling paradigm conflict in artistic text image synthesis, and the lack of text structure semantic understanding leads to low image data synthesis accuracy is solved, and the accuracy of image data synthesis is improved.

Images