Buckets:
| # ControlNet | |
| [ControlNet](https://huggingface.co/papers/2302.05543) dirige un modelo de difusión preentrenado mediante una imagen de control estructural (bordes, profundidad o pose) mientras se mantiene un prompt de texto. Congela el modelo base y añade una red paralela cuyos residuos guían al desruidador, lo que permite controlar el diseño sin necesidad de afinar el modelo. | |
| Carga un ControlNet para el control que necesites (por ejemplo, canny) y pásalo como `controlnet=` a [from_pretrained()](/docs/diffusers/main/es/api/pipelines/overview#diffusers.DiffusionPipeline.from_pretrained). Usa `controlnet_conditioning_scale` para establecer la fuerza con la que el control dirige la generación. | |
| > [!TIP] | |
| > ControlNets están disponibles para muchos modelos, como [Flux](../api/pipelines/controlnet_flux), [Hunyuan-DiT](../api/pipelines/controlnet_hunyuandit), [Stable Diffusion 3](../api/pipelines/controlnet_sd3) y más. Los ejemplos de esta guía utilizan Flux y Stable Diffusion XL. Para T2I-Adapter en puntos de control antiguos de la familia Stable Diffusion, consulte [Adaptadores heredados](./legacy_adapters#t2i-adapter). Prefiera ControlNet para nuevo trabajo de generación controlable. | |
| Las pestañas de abajo muestran ControlNet para texto a imagen, imagen a imagen y relleno (inpainting). | |
| Genera una imagen canny con [opencv-python](https://github.com/opencv/opencv-python). | |
| ```py | |
| import cv2 | |
| import numpy as np | |
| from PIL import Image | |
| from diffusers.utils import load_image | |
| original_image = load_image( | |
| "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/non-enhanced-prompt.png" | |
| ) | |
| image = np.array(original_image) | |
| low_threshold = 100 | |
| high_threshold = 200 | |
| image = cv2.Canny(image, low_threshold, high_threshold) | |
| image = image[:, :, None] | |
| image = np.concatenate([image, image, image], axis=2) | |
| canny_image = Image.fromarray(image) | |
| ``` | |
| Pasa la imagen canny a la pipeline. Usa el parámetro `controlnet_conditioning_scale` para determinar cuánto peso asignar al control. | |
| ```py | |
| import torch | |
| from diffusers.utils import load_image | |
| from diffusers import FluxControlNetPipeline, FluxControlNetModel | |
| controlnet = FluxControlNetModel.from_pretrained( | |
| "InstantX/FLUX.1-dev-Controlnet-Canny", dtype=torch.bfloat16 | |
| ) | |
| pipeline = FluxControlNetPipeline.from_pretrained( | |
| "black-forest-labs/FLUX.1-dev", controlnet=controlnet, dtype=torch.bfloat16 | |
| ).to("cuda") # or "mps", "xpu", "cpu" | |
| prompt = """ | |
| A photorealistic overhead image of a cat reclining sideways in a flamingo pool floatie holding a margarita. | |
| The cat is floating leisurely in the pool and completely relaxed and happy. | |
| """ | |
| pipeline( | |
| prompt, | |
| control_image=canny_image, | |
| controlnet_conditioning_scale=0.5, | |
| num_inference_steps=50, | |
| guidance_scale=3.5, | |
| ).images[0] | |
| ``` | |
| imagen original | |
| imagen canny | |
| imagen generada | |
| Genera un mapa de profundidad con un pipeline de estimación de profundidad de Transformers. | |
| ```py | |
| import torch | |
| import numpy as np | |
| from PIL import Image | |
| from transformers import DPTImageProcessor, DPTForDepthEstimation | |
| from diffusers import ControlNetModel, StableDiffusionXLControlNetImg2ImgPipeline, AutoencoderKL | |
| from diffusers.utils import load_image | |
| device = "cuda" # or "mps", "xpu", "cpu" | |
| depth_estimator = DPTForDepthEstimation.from_pretrained("Intel/dpt-hybrid-midas").to(device) | |
| feature_extractor = DPTImageProcessor.from_pretrained("Intel/dpt-hybrid-midas") | |
| def get_depth_map(image): | |
| image = feature_extractor(images=image, return_tensors="pt").pixel_values.to(device) | |
| with torch.no_grad(), torch.autocast(device): | |
| depth_map = depth_estimator(image).predicted_depth | |
| depth_map = torch.nn.functional.interpolate( | |
| depth_map.unsqueeze(1), | |
| size=(1024, 1024), | |
| mode="bicubic", | |
| align_corners=False, | |
| ) | |
| depth_min = torch.amin(depth_map, dim=[1, 2, 3], keepdim=True) | |
| depth_max = torch.amax(depth_map, dim=[1, 2, 3], keepdim=True) | |
| depth_map = (depth_map - depth_min) / (depth_max - depth_min) | |
| image = torch.cat([depth_map] * 3, dim=1) | |
| image = image.permute(0, 2, 3, 1).cpu().numpy()[0] | |
| image = Image.fromarray((image * 255.0).clip(0, 255).astype(np.uint8)) | |
| return image | |
| image = load_image( | |
| "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/non-enhanced-prompt.png" | |
| ).resize((1024, 1024)) | |
| depth_image = get_depth_map(image) | |
| ``` | |
| Pasa el mapa de profundidad a la pipeline. Usa el parámetro `controlnet_conditioning_scale` para determinar cuánto peso asignar al control. | |
| ```py | |
| controlnet = ControlNetModel.from_pretrained( | |
| "diffusers/controlnet-depth-sdxl-1.0-small", | |
| dtype=torch.float16, | |
| ) | |
| vae = AutoencoderKL.from_pretrained("madebyollin/sdxl-vae-fp16-fix", dtype=torch.float16) | |
| pipeline = StableDiffusionXLControlNetImg2ImgPipeline.from_pretrained( | |
| "stabilityai/stable-diffusion-xl-base-1.0", | |
| controlnet=controlnet, | |
| vae=vae, | |
| dtype=torch.float16, | |
| ).to(device) | |
| prompt = """ | |
| A photorealistic overhead image of a cat reclining sideways in a flamingo pool floatie holding a margarita. | |
| The cat is floating leisurely in the pool and completely relaxed and happy. | |
| """ | |
| controlnet_conditioning_scale = 0.5 | |
| pipeline( | |
| prompt, | |
| image=image, | |
| control_image=depth_image, | |
| controlnet_conditioning_scale=controlnet_conditioning_scale, | |
| strength=0.99, | |
| num_inference_steps=100, | |
| ).images[0] | |
| ``` | |
| imagen original | |
| mapa de profundidad | |
| imagen generada | |
| Genere una imagen de máscara que marque qué píxeles reemplazar y pásela como `mask_image` junto con la imagen de control. | |
| ```py | |
| import cv2 | |
| import torch | |
| import numpy as np | |
| from PIL import Image | |
| from diffusers.utils import load_image | |
| from diffusers import StableDiffusionXLControlNetInpaintPipeline, ControlNetModel | |
| init_image = load_image( | |
| "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/non-enhanced-prompt.png" | |
| ) | |
| init_image = init_image.resize((1024, 1024)) | |
| mask_image = load_image( | |
| "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/cat_mask.png" | |
| ) | |
| mask_image = mask_image.resize((1024, 1024)) | |
| def make_canny_condition(image): | |
| image = np.array(image) | |
| image = cv2.Canny(image, 100, 200) | |
| image = image[:, :, None] | |
| image = np.concatenate([image, image, image], axis=2) | |
| image = Image.fromarray(image) | |
| return image | |
| control_image = make_canny_condition(init_image) | |
| ``` | |
| Pasa la máscara y la imagen de control a la pipeline. Usa el parámetro `controlnet_conditioning_scale` para determinar cuánto peso asignar al control. | |
| ```py | |
| controlnet = ControlNetModel.from_pretrained( | |
| "diffusers/controlnet-canny-sdxl-1.0", dtype=torch.float16 | |
| ) | |
| pipeline = StableDiffusionXLControlNetInpaintPipeline.from_pretrained( | |
| "stabilityai/stable-diffusion-xl-base-1.0", controlnet=controlnet, dtype=torch.float16 | |
| ).to("cuda") # or "mps", "xpu", "cpu" | |
| pipeline( | |
| "a cute and fluffy bunny rabbit", | |
| num_inference_steps=100, | |
| strength=0.99, | |
| controlnet_conditioning_scale=0.5, | |
| image=init_image, | |
| mask_image=mask_image, | |
| control_image=control_image, | |
| ).images[0] | |
| ``` | |
| imagen original | |
| imagen de máscara | |
| imagen generada | |
| ## Multi-ControlNet | |
| Compón varios ControlNets (por ejemplo, canny + depth) pasándolos como una lista. Enmascara las regiones superpuestas cuando sea posible y ajusta cada `controlnet_conditioning_scale`. Crea `canny_image` con OpenCV Canny (como en text-to-image) y `depth_image` con DPT (como en image-to-image). Redimensiona cada imagen de control al tamaño del pipeline antes de llamar al pipeline. | |
| ```py | |
| import cv2 | |
| import torch | |
| import numpy as np | |
| from PIL import Image | |
| from transformers import DPTImageProcessor, DPTForDepthEstimation | |
| from diffusers import StableDiffusionXLControlNetPipeline, ControlNetModel, AutoencoderKL | |
| from diffusers.utils import load_image | |
| original_image = load_image( | |
| "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/non-enhanced-prompt.png" | |
| ) | |
| image = np.array(original_image) | |
| canny_image = Image.fromarray( | |
| np.concatenate([cv2.Canny(image, 100, 200)[:, :, None]] * 3, axis=2) | |
| ) | |
| device = "cuda" # or "mps", "xpu", "cpu" | |
| depth_estimator = DPTForDepthEstimation.from_pretrained("Intel/dpt-hybrid-midas").to(device) | |
| feature_extractor = DPTImageProcessor.from_pretrained("Intel/dpt-hybrid-midas") | |
| def get_depth_map(image): | |
| image = feature_extractor(images=image, return_tensors="pt").pixel_values.to(device) | |
| with torch.no_grad(), torch.autocast(device): | |
| depth_map = depth_estimator(image).predicted_depth | |
| depth_map = torch.nn.functional.interpolate( | |
| depth_map.unsqueeze(1), | |
| size=(1024, 1024), | |
| mode="bicubic", | |
| align_corners=False, | |
| ) | |
| depth_min = torch.amin(depth_map, dim=[1, 2, 3], keepdim=True) | |
| depth_max = torch.amax(depth_map, dim=[1, 2, 3], keepdim=True) | |
| depth_map = (depth_map - depth_min) / (depth_max - depth_min) | |
| image = torch.cat([depth_map] * 3, dim=1) | |
| image = image.permute(0, 2, 3, 1).cpu().numpy()[0] | |
| image = Image.fromarray((image * 255.0).clip(0, 255).astype(np.uint8)) | |
| return image | |
| depth_image = get_depth_map(original_image.resize((1024, 1024))) | |
| controlnets = [ | |
| ControlNetModel.from_pretrained( | |
| "diffusers/controlnet-canny-sdxl-1.0", dtype=torch.float16 | |
| ), | |
| ControlNetModel.from_pretrained( | |
| "diffusers/controlnet-depth-sdxl-1.0-small", dtype=torch.float16 | |
| ), | |
| ] | |
| vae = AutoencoderKL.from_pretrained("madebyollin/sdxl-vae-fp16-fix", dtype=torch.float16) | |
| pipeline = StableDiffusionXLControlNetPipeline.from_pretrained( | |
| "stabilityai/stable-diffusion-xl-base-1.0", controlnet=controlnets, vae=vae, dtype=torch.float16 | |
| ).to(device) | |
| prompt = """ | |
| a relaxed rabbit sitting on a striped towel next to a pool with a tropical drink nearby, | |
| bright sunny day, vacation scene, 35mm photograph, film, professional, 4k, highly detailed | |
| """ | |
| negative_prompt = "lowres, bad anatomy, worst quality, low quality, deformed, ugly" | |
| images = [canny_image.resize((1024, 1024)), depth_image.resize((1024, 1024))] | |
| pipeline( | |
| prompt, | |
| negative_prompt=negative_prompt, | |
| image=images, | |
| num_inference_steps=100, | |
| controlnet_conditioning_scale=[0.5, 0.5], | |
| ).images[0] | |
| ``` | |
| imagen canny | |
| mapa de profundidad | |
| imagen generada | |
| ## Modo de conjetura | |
| [Modo de conjetura](https://github.com/lllyasviel/ControlNet/discussions/188) genera una imagen a partir de un control por sí solo cuando no se dispone de un prompt de texto (o no se desea usar uno). Escala los residuales de ControlNet según la profundidad del bloque, desde aproximadamente `0.1` en los bloques de bajada iniciales hasta `1.0` en el bloque intermedio. Esto evita que las capas iniciales sobre-restringan la generación, mientras que los bloques más profundos siguen aportando la estructura. | |
| ```py | |
| import torch | |
| from diffusers.utils import load_image | |
| from diffusers import StableDiffusionXLControlNetPipeline, ControlNetModel | |
| controlnet = ControlNetModel.from_pretrained( | |
| "diffusers/controlnet-canny-sdxl-1.0", dtype=torch.float16 | |
| ) | |
| pipeline = StableDiffusionXLControlNetPipeline.from_pretrained( | |
| "stabilityai/stable-diffusion-xl-base-1.0", | |
| controlnet=controlnet, | |
| dtype=torch.float16 | |
| ).to("cuda") # or "mps", "xpu", "cpu" | |
| canny_image = load_image("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/canny-cat.png") | |
| pipeline( | |
| "", | |
| image=canny_image, | |
| guess_mode=True | |
| ).images[0] | |
| ``` | |
| imagen canny | |
| imagen generada | |
Xet Storage Details
- Size:
- 11.8 kB
- Xet hash:
- 90f7899816405fa8010a92c1815614482790f8a2cac621bf268a6ac5a0dbe060
·
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.