Each character must replicate exactly the same facial expression from the input images Stable Diffusion SDXL 1.0 prompts

no results