许可协议：MIT
库名称：diffusers
管道标签：image-to-image

DiLightNet：基于扩散模型的图像生成细粒度光照控制系统

SIGGRAPH 2024

• 项目主页：https://dilightnet.github.io/
• 论文：https://arxiv.org/abs/2402.11929
• 完整用法：请查看 https://github.com/iamNCJ/DiLightNet

使用示例：

from diffusers.utils import get_class_from_dynamic_module  
NeuralTextureControlNetModel = get_class_from_dynamic_module(  
    "dilightnet/model_helpers",  
    "neuraltexture_controlnet.py",  
    "NeuralTextureControlNetModel"  
)  
neuraltexture_controlnet = NeuralTextureControlNetModel.from_pretrained("DiLightNet/DiLightNet")  

pipe = StableDiffusionControlNetPipeline.from_pretrained(  
    "stabilityai/stable-diffusion-2-1", controlnet=neuraltexture_controlnet,  
)  
cond_image = torch.randn((1, 16, 512, 512))  
image = pipe("输入文本提示", image=cond_image).images[0]