A method and apparatus for generating digital human images based on template-driven and pose alignment

Abstract

This invention relates to a method and apparatus for generating digital human images based on template-driven and pose alignment. The method includes: acquiring original images and fixed templates, and calling text set parameters; an image generation module preprocessing the original images and inputting them into an image-generated model to generate stylized static images; a pose alignment module providing compliant pose sequences for generating complex action videos based on visual action tags in a preset action template library or user-uploaded action videos; a micro-video generation module generating fixed-duration micro-videos through micro-action-driven or pose-driven methods; and a dynamic-static loop splicing module optimizing video continuity through a loop strategy to achieve dynamic switching between standby and interactive states. This invention integrates template control, image generation, video synthesis, pose alignment, and loop splicing to construct a complete digital human generation solution covering static to dynamic, preset to customized, and generated to playback processes.

Images