Subject Reference-To-Video

Aims to create videos featuring a specific, consistent subject (**only face of human) based on provided reference image and a descriptive text prompt.