ComfyUI预处理器ControlNet简单介绍与使用（附件工作流）

已于 2024-07-07 14:19:07 修改
阅读量608 收藏 4
点赞数 7
分类专栏： ComfyUI 文章标签： stable diffusion
于 2024-07-07 14:16:31 首次发布
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/weixin_38468167/article/details/140245417
版权

简介

ControlNet 是一个很强的插件，提供了很多种图片的控制方式，有的可以控制画面的结构，有的可以控制人物的姿势，还有的可以控制图片的画风，这对于提高AI绘画的质量特别有用。接下来就演示几种热门常用的控制方式

1.OpenPose（姿态控制预处理器）

姿态控制预处理器可以根据提供的图像将人物的骨骼脸部手部的姿态展示处理，通过这个预处理器可以很好的控制出图人物的姿态

2.Depth（深度预处理器）

深度预处理器可以将图片的空间的远近以黑白的形式展示出来，白近黑远，当我们上传一张图片通过OpenPose识别到手的位置，但骨骼图并不能描述手在身前还是身后的时候，那个深度预处理器就可以提现出作用了，当然还可以运用在一些建筑、室内等情况

3.LineArt（线条预处理器）

线条预处理器可以将图片用线条的形式描绘出来，可以很好的控制图片的细节

4.HED Soft-Edge（模糊线条预处理器）

模糊线条预处理器与线条预处理器类型也是用线条描绘图片，但仅大概描绘轮廓，更利于出图的随机性

接下来演示一下这四个预处理器效果,不同的预处理器之间是可以搭配使用的，根据不同的需求选择不用的预处理器来解决问题

请添加图片描述

附件工作流

复制Json内容到ComfyUI中即可

{
  "last_node_id": 46,
  "last_link_id": 65,
  "nodes": [
    {
      "id": 31,
      "type": "PreviewImage",
      "pos": [
        1110,
        -1000
      ],
      "size": {
        "0": 210,
        "1": 310
      },
      "flags": {},
      "order": 15,
      "mode": 0,
      "inputs": [
        {
          "name": "images",
          "type": "IMAGE",
          "link": 31,
          "label": "图像"
        }
      ],
      "properties": {
        "Node name for S&R": "PreviewImage"
      }
    },
    {
      "id": 29,
      "type": "PreviewImage",
      "pos": [
        760,
        -1000
      ],
      "size": {
        "0": 210,
        "1": 310
      },
      "flags": {},
      "order": 14,
      "mode": 0,
      "inputs": [
        {
          "name": "images",
          "type": "IMAGE",
          "link": 30,
          "label": "图像",
          "slot_index": 0
        }
      ],
      "properties": {
        "Node name for S&R": "PreviewImage"
      }
    },
    {
      "id": 34,
      "type": "HEDPreprocessor",
      "pos": [
        1050,
        -500
      ],
      "size": {
        "0": 315,
        "1": 82
      },
      "flags": {},
      "order": 10,
      "mode": 0,
      "inputs": [
        {
          "name": "image",
          "type": "IMAGE",
          "link": 34,
          "label": "图像"
        }
      ],
      "outputs": [
        {
          "name": "IMAGE",
          "type": "IMAGE",
          "links": [
            31,
            54
          ],
          "shape": 3,
          "label": "图像",
          "slot_index": 0
        }
      ],
      "properties": {
        "Node name for S&R": "HEDPreprocessor"
      },
      "widgets_values": [
        "enable",
        512
      ]
    },
    {
      "id": 32,
      "type": "ControlNetLoader",
      "pos": [
        1060,
        -640
      ],
      "size": {
        "0": 315,
        "1": 58
      },
      "flags": {},
      "order": 0,
      "mode": 0,
      "outputs": [
        {
          "name": "CONTROL_NET",
          "type": "CONTROL_NET",
          "links": [
            53
          ],
          "shape": 3,
          "label": "ControlNet",
          "slot_index": 0
        }
      ],
      "properties": {
        "Node name for S&R": "ControlNetLoader"
      },
      "widgets_values": [
        "control_v11p_sd15_softedge.pth"
      ]
    },
    {
      "id": 25,
      "type": "PreviewImage",
      "pos": [
        420,
        -1000
      ],
      "size": {
        "0": 210,
        "1": 310
      },
      "flags": {},
      "order": 13,
      "mode": 0,
      "inputs": [
        {
          "name": "images",
          "type": "IMAGE",
          "link": 29,
          "label": "图像"
        }
      ],
      "properties": {
        "Node name for S&R": "PreviewImage"
      }
    },
    {
      "id": 12,
      "type": "ControlNetLoader",
      "pos": [
        30,
        -640
      ],
      "size": {
        "0": 320,
        "1": 60
      },
      "flags": {},
      "order": 1,
      "mode": 0,
      "outputs": [
        {
          "name": "CONTROL_NET",
          "type": "CONTROL_NET",
          "links": [
            47
          ],
          "slot_index": 0,
          "label": "ControlNet"
        }
      ],
      "properties": {
        "Node name for S&R": "ControlNetLoader"
      },
      "widgets_values": [
        "control_v11p_sd15_openpose.pth"
      ]
    },
    {
      "id": 16,
      "type": "OpenposePreprocessor",
      "pos": [
        30,
        -540
      ],
      "size": {
        "0": 315,
        "1": 150
      },
      "flags": {},
      "order": 7,
      "mode": 0,
      "inputs": [
        {
          "name": "image",
          "type": "IMAGE",
          "link": 22,
          "label": "图像"
        }
      ],
      "outputs": [
        {
          "name": "IMAGE",
          "type": "IMAGE",
          "links": [
            24,
            48
          ],
          "shape": 3,
          "label": "图像",
          "slot_index": 0
        },
        {
          "name": "POSE_KEYPOINT",
          "type": "POSE_KEYPOINT",
          "links": null,
          "shape": 3,
          "label": "姿态关键点"
        }
      ],
      "properties": {
        "Node name for S&R": "OpenposePreprocessor"
      },
      "widgets_values": [
        "enable",
        "enable",
        "enable",
        512
      ]
    },
    {
      "id": 23,
      "type": "ControlNetLoader",
      "pos": [
        380,
        -640
      ],
      "size": {
        "0": 315,
        "1": 58
      },
      "flags": {},
      "order": 2,
      "mode": 0,
      "outputs": [
        {
          "name": "CONTROL_NET",
          "type": "CONTROL_NET",
          "links": [
            49
          ],
          "shape": 3,
          "label": "ControlNet",
          "slot_index": 0
        }
      ],
      "properties": {
        "Node name for S&R": "ControlNetLoader"
      },
      "widgets_values": [
        "control_v11f1p_sd15_depth.pth"
      ]
    },
    {
      "id": 21,
      "type": "Zoe_DepthAnythingPreprocessor",
      "pos": [
        370,
        -500
      ],
      "size": {
        "0": 315,
        "1": 82
      },
      "flags": {},
      "order": 8,
      "mode": 0,
      "inputs": [
        {
          "name": "image",
          "type": "IMAGE",
          "link": 32,
          "label": "图像",
          "slot_index": 0
        }
      ],
      "outputs": [
        {
          "name": "IMAGE",
          "type": "IMAGE",
          "links": [
            29,
            50
          ],
          "shape": 3,
          "label": "图像",
          "slot_index": 0
        }
      ],
      "properties": {
        "Node name for S&R": "Zoe_DepthAnythingPreprocessor"
      },
      "widgets_values": [
        "indoor",
        512
      ]
    },
    {
      "id": 30,
      "type": "ControlNetLoader",
      "pos": [
        710,
        -640
      ],
      "size": {
        "0": 315,
        "1": 58
      },
      "flags": {},
      "order": 3,
      "mode": 0,
      "outputs": [
        {
          "name": "CONTROL_NET",
          "type": "CONTROL_NET",
          "links": [
            51
          ],
          "shape": 3,
          "label": "ControlNet",
          "slot_index": 0
        }
      ],
      "properties": {
        "Node name for S&R": "ControlNetLoader"
      },
      "widgets_values": [
        "control_v11p_sd15_lineart.pth"
      ]
    },
    {
      "id": 33,
      "type": "LineArtPreprocessor",
      "pos": [
        710,
        -500
      ],
      "size": {
        "0": 315,
        "1": 82
      },
      "flags": {},
      "order": 9,
      "mode": 0,
      "inputs": [
        {
          "name": "image",
          "type": "IMAGE",
          "link": 33,
          "label": "图像"
        }
      ],
      "outputs": [
        {
          "name": "IMAGE",
          "type": "IMAGE",
          "links": [
            30,
            52
          ],
          "shape": 3,
          "label": "图像",
          "slot_index": 0
        }
      ],
      "properties": {
        "Node name for S&R": "LineArtPreprocessor"
      },
      "widgets_values": [
        "disable",
        512
      ]
    },
    {
      "id": 35,
      "type": "Efficient Loader",
      "pos": [
        2130,
        690
      ],
      "size": {
        "0": 400,
        "1": 462
      },
      "flags": {},
      "order": 6,
      "mode": 0,
      "inputs": [
        {
          "name": "lora_stack",
          "type": "LORA_STACK",
          "link": null,
          "label": "LoRA堆"
        },
        {
          "name": "cnet_stack",
          "type": "CONTROL_NET_STACK",
          "link": 55,
          "label": "ControlNet堆",
          "slot_index": 1
        }
      ],
      "outputs": [
        {
          "name": "MODEL",
          "type": "MODEL",
          "links": null,
          "shape": 3,
          "label": "模型"
        },
        {
          "name": "CONDITIONING+",
          "type": "CONDITIONING",
          "links": null,
          "shape": 3,
          "label": "正面条件"
        },
        {
          "name": "CONDITIONING-",
          "type": "CONDITIONING",
          "links": null,
          "shape": 3,
          "label": "负面条件"
        },
        {
          "name": "LATENT",
          "type": "LATENT",
          "links": null,
          "shape": 3,
          "label": "Latent"
        },
        {
          "name": "VAE",
          "type": "VAE",
          "links": null,
          "shape": 3,
          "label": "VAE"
        },
        {
          "name": "CLIP",
          "type": "CLIP",
          "links": null,
          "shape": 3,
          "label": "CLIP"
        },
        {
          "name": "DEPENDENCIES",
          "type": "DEPENDENCIES",
          "links": null,
          "shape": 3,
          "label": "依赖"
        }
      ],
      "properties": {
        "Node name for S&R": "Efficient Loader"
      },
      "widgets_values": [
        "AWPainting_v1.3.safetensors",
        "Baked VAE",
        -1,
        "None",
        1,
        1,
        "CLIP_POSITIVE",
        "CLIP_NEGATIVE",
        "none",
        "comfy",
        512,
        512,
        1
      ],
      "color": "#2a363b",
      "bgcolor": "#3f5159",
      "shape": 1
    },
    {
      "id": 41,
      "type": "Control Net Stacker",
      "pos": [
        1720,
        650
      ],
      "size": {
        "0": 315,
        "1": 146
      },
      "flags": {},
      "order": 4,
      "mode": 0,
      "inputs": [
        {
          "name": "control_net",
          "type": "CONTROL_NET",
          "link": null,
          "label": "ControlNet"
        },
        {
          "name": "image",
          "type": "IMAGE",
          "link": null,
          "label": "图像"
        },
        {
          "name": "cnet_stack",
          "type": "CONTROL_NET_STACK",
          "link": null,
          "label": "ControlNet堆"
        }
      ],
      "outputs": [
        {
          "name": "CNET_STACK",
          "type": "CONTROL_NET_STACK",
          "links": [
            55
          ],
          "shape": 3,
          "label": "ControlNet堆"
        }
      ],
      "properties": {
        "Node name for S&R": "Control Net Stacker"
      },
      "widgets_values": [
        1,
        0,
        1
      ],
      "color": "#223322",
      "bgcolor": "#335533",
      "shape": 1
    },
    {
      "id": 17,
      "type": "PreviewImage",
      "pos": [
        70,
        -1000
      ],
      "size": {
        "0": 210,
        "1": 310
      },
      "flags": {},
      "order": 11,
      "mode": 0,
      "inputs": [
        {
          "name": "images",
          "type": "IMAGE",
          "link": 24,
          "label": "图像"
        }
      ],
      "properties": {
        "Node name for S&R": "PreviewImage"
      }
    },
    {
      "id": 37,
      "type": "Control Net Stacker",
      "pos": [
        30,
        -340
      ],
      "size": {
        "0": 315,
        "1": 146
      },
      "flags": {},
      "order": 12,
      "mode": 0,
      "inputs": [
        {
          "name": "control_net",
          "type": "CONTROL_NET",
          "link": 47,
          "label": "ControlNet"
        },
        {
          "name": "image",
          "type": "IMAGE",
          "link": 48,
          "label": "图像"
        },
        {
          "name": "cnet_stack",
          "type": "CONTROL_NET_STACK",
          "link": null,
          "label": "ControlNet堆"
        }
      ],
      "outputs": [
        {
          "name": "CNET_STACK",
          "type": "CONTROL_NET_STACK",
          "links": [
            62
          ],
          "shape": 3,
          "label": "ControlNet堆",
          "slot_index": 0
        }
      ],
      "properties": {
        "Node name for S&R": "Control Net Stacker"
      },
      "widgets_values": [
        1,
        0,
        1
      ],
      "color": "#223322",
      "bgcolor": "#335533",
      "shape": 1
    },
    {
      "id": 38,
      "type": "Control Net Stacker",
      "pos": [
        370,
        -340
      ],
      "size": {
        "0": 315,
        "1": 146
      },
      "flags": {},
      "order": 16,
      "mode": 0,
      "inputs": [
        {
          "name": "control_net",
          "type": "CONTROL_NET",
          "link": 49,
          "label": "ControlNet"
        },
        {
          "name": "image",
          "type": "IMAGE",
          "link": 50,
          "label": "图像"
        },
        {
          "name": "cnet_stack",
          "type": "CONTROL_NET_STACK",
          "link": 62,
          "label": "ControlNet堆"
        }
      ],
      "outputs": [
        {
          "name": "CNET_STACK",
          "type": "CONTROL_NET_STACK",
          "links": [
            63
          ],
          "shape": 3,
          "label": "ControlNet堆",
          "slot_index": 0
        }
      ],
      "properties": {
        "Node name for S&R": "Control Net Stacker"
      },
      "widgets_values": [
        0.2,
        0,
        1
      ],
      "color": "#223322",
      "bgcolor": "#335533",
      "shape": 1
    },
    {
      "id": 39,
      "type": "Control Net Stacker",
      "pos": [
        710,
        -340
      ],
      "size": {
        "0": 315,
        "1": 146
      },
      "flags": {},
      "order": 17,
      "mode": 0,
      "inputs": [
        {
          "name": "control_net",
          "type": "CONTROL_NET",
          "link": 51,
          "label": "ControlNet"
        },
        {
          "name": "image",
          "type": "IMAGE",
          "link": 52,
          "label": "图像"
        },
        {
          "name": "cnet_stack",
          "type": "CONTROL_NET_STACK",
          "link": 63,
          "label": "ControlNet堆"
        }
      ],
      "outputs": [
        {
          "name": "CNET_STACK",
          "type": "CONTROL_NET_STACK",
          "links": [
            64
          ],
          "shape": 3,
          "label": "ControlNet堆",
          "slot_index": 0
        }
      ],
      "properties": {
        "Node name for S&R": "Control Net Stacker"
      },
      "widgets_values": [
        0.5,
        0,
        1
      ],
      "color": "#223322",
      "bgcolor": "#335533",
      "shape": 1
    },
    {
      "id": 40,
      "type": "Control Net Stacker",
      "pos": [
        1050,
        -340
      ],
      "size": {
        "0": 315,
        "1": 146
      },
      "flags": {},
      "order": 18,
      "mode": 0,
      "inputs": [
        {
          "name": "control_net",
          "type": "CONTROL_NET",
          "link": 53,
          "label": "ControlNet",
          "slot_index": 0
        },
        {
          "name": "image",
          "type": "IMAGE",
          "link": 54,
          "label": "图像",
          "slot_index": 1
        },
        {
          "name": "cnet_stack",
          "type": "CONTROL_NET_STACK",
          "link": 64,
          "label": "ControlNet堆"
        }
      ],
      "outputs": [
        {
          "name": "CNET_STACK",
          "type": "CONTROL_NET_STACK",
          "links": [
            65
          ],
          "shape": 3,
          "label": "ControlNet堆",
          "slot_index": 0
        }
      ],
      "properties": {
        "Node name for S&R": "Control Net Stacker"
      },
      "widgets_values": [
        0.5,
        0,
        1
      ],
      "color": "#223322",
      "bgcolor": "#335533",
      "shape": 1
    },
    {
      "id": 44,
      "type": "SaveImage",
      "pos": [
        2220,
        -630
      ],
      "size": {
        "0": 320,
        "1": 270
      },
      "flags": {},
      "order": 21,
      "mode": 0,
      "inputs": [
        {
          "name": "images",
          "type": "IMAGE",
          "link": 61,
          "label": "图像"
        }
      ],
      "properties": {},
      "widgets_values": [
        "ComfyUI"
      ]
    },
    {
      "id": 43,
      "type": "KSampler (Efficient)",
      "pos": [
        1860,
        -750
      ],
      "size": {
        "0": 330,
        "1": 560
      },
      "flags": {},
      "order": 20,
      "mode": 0,
      "inputs": [
        {
          "name": "model",
          "type": "MODEL",
          "link": 56,
          "label": "模型"
        },
        {
          "name": "positive",
          "type": "CONDITIONING",
          "link": 57,
          "label": "正面条件"
        },
        {
          "name": "negative",
          "type": "CONDITIONING",
          "link": 58,
          "label": "负面条件"
        },
        {
          "name": "latent_image",
          "type": "LATENT",
          "link": 59,
          "label": "Latent"
        },
        {
          "name": "optional_vae",
          "type": "VAE",
          "link": 60,
          "label": "VAE(可选)"
        },
        {
          "name": "script",
          "type": "SCRIPT",
          "link": null,
          "label": "脚本"
        }
      ],
      "outputs": [
        {
          "name": "MODEL",
          "type": "MODEL",
          "links": null,
          "shape": 3,
          "label": "模型"
        },
        {
          "name": "CONDITIONING+",
          "type": "CONDITIONING",
          "links": null,
          "shape": 3,
          "label": "正面条件"
        },
        {
          "name": "CONDITIONING-",
          "type": "CONDITIONING",
          "links": null,
          "shape": 3,
          "label": "负面条件"
        },
        {
          "name": "LATENT",
          "type": "LATENT",
          "links": null,
          "shape": 3,
          "label": "Latent"
        },
        {
          "name": "VAE",
          "type": "VAE",
          "links": null,
          "shape": 3,
          "label": "VAE"
        },
        {
          "name": "IMAGE",
          "type": "IMAGE",
          "links": [
            61
          ],
          "shape": 3,
          "label": "图像",
          "slot_index": 5
        }
      ],
      "properties": {
        "Node name for S&R": "KSampler (Efficient)"
      },
      "widgets_values": [
        248390093388796,
        null,
        20,
        10,
        "dpmpp_2m",
        "karras",
        1,
        "auto",
        "true"
      ],
      "color": "#222233",
      "bgcolor": "#333355",
      "shape": 1
    },
    {
      "id": 42,
      "type": "Efficient Loader",
      "pos": [
        1420,
        -650
      ],
      "size": {
        "0": 400,
        "1": 462.0000305175781
      },
      "flags": {},
      "order": 19,
      "mode": 0,
      "inputs": [
        {
          "name": "lora_stack",
          "type": "LORA_STACK",
          "link": null,
          "label": "LoRA堆"
        },
        {
          "name": "cnet_stack",
          "type": "CONTROL_NET_STACK",
          "link": 65,
          "label": "ControlNet堆"
        }
      ],
      "outputs": [
        {
          "name": "MODEL",
          "type": "MODEL",
          "links": [
            56
          ],
          "shape": 3,
          "label": "模型",
          "slot_index": 0
        },
        {
          "name": "CONDITIONING+",
          "type": "CONDITIONING",
          "links": [
            57
          ],
          "shape": 3,
          "label": "正面条件",
          "slot_index": 1
        },
        {
          "name": "CONDITIONING-",
          "type": "CONDITIONING",
          "links": [
            58
          ],
          "shape": 3,
          "label": "负面条件",
          "slot_index": 2
        },
        {
          "name": "LATENT",
          "type": "LATENT",
          "links": [
            59
          ],
          "shape": 3,
          "label": "Latent",
          "slot_index": 3
        },
        {
          "name": "VAE",
          "type": "VAE",
          "links": [
            60
          ],
          "shape": 3,
          "label": "VAE",
          "slot_index": 4
        },
        {
          "name": "CLIP",
          "type": "CLIP",
          "links": null,
          "shape": 3,
          "label": "CLIP"
        },
        {
          "name": "DEPENDENCIES",
          "type": "DEPENDENCIES",
          "links": null,
          "shape": 3,
          "label": "依赖"
        }
      ],
      "properties": {
        "Node name for S&R": "Efficient Loader"
      },
      "widgets_values": [
        "AWPainting_v1.3.safetensors",
        "Baked VAE",
        -2,
        "None",
        1,
        1,
        "masterpiece, best quality, highres, 1girl, bare shoulders, brown hair, long hair, (orange dress:1.2), looking at viewer, forest, maple leaves,outdoors, wild, plants, cinematic lights, lightrays,depth of field, blurry_background, blurry_foreground, shiny luminious,",
        "(hands), text, error, cropped, (worst quality:1.2), (low quality:1.2), normal quality, (jpeg artifacts:1.3), signature, watermark, username, blurry, artist name, monochrome, sketch, censorship, censor, (copyright:1.2), extra legs, (forehead mark) (depth of field) (emotionless) (penis), embedding:EasyNegative, embedding:badhandv4, ",
        "none",
        "comfy++",
        768,
        1152,
        1
      ],
      "color": "#443322",
      "bgcolor": "#665533",
      "shape": 1
    },
    {
      "id": 11,
      "type": "LoadImage",
      "pos": [
        -250,
        -800
      ],
      "size": [
        220,
        314
      ],
      "flags": {},
      "order": 5,
      "mode": 0,
      "outputs": [
        {
          "name": "IMAGE",
          "type": "IMAGE",
          "links": [
            22,
            32,
            33,
            34
          ],
          "slot_index": 0,
          "label": "图像"
        },
        {
          "name": "MASK",
          "type": "MASK",
          "links": null,
          "label": "遮罩"
        }
      ],
      "properties": {
        "Node name for S&R": "LoadImage"
      },
      "widgets_values": [
        "aimake_Example_1713001562048_49 (1) (3).jpg",
        "image"
      ]
    }
  ],
  "links": [
    [
      22,
      11,
      0,
      16,
      0,
      "IMAGE"
    ],
    [
      24,
      16,
      0,
      17,
      0,
      "IMAGE"
    ],
    [
      29,
      21,
      0,
      25,
      0,
      "IMAGE"
    ],
    [
      30,
      33,
      0,
      29,
      0,
      "IMAGE"
    ],
    [
      31,
      34,
      0,
      31,
      0,
      "IMAGE"
    ],
    [
      32,
      11,
      0,
      21,
      0,
      "IMAGE"
    ],
    [
      33,
      11,
      0,
      33,
      0,
      "IMAGE"
    ],
    [
      34,
      11,
      0,
      34,
      0,
      "IMAGE"
    ],
    [
      47,
      12,
      0,
      37,
      0,
      "CONTROL_NET"
    ],
    [
      48,
      16,
      0,
      37,
      1,
      "IMAGE"
    ],
    [
      49,
      23,
      0,
      38,
      0,
      "CONTROL_NET"
    ],
    [
      50,
      21,
      0,
      38,
      1,
      "IMAGE"
    ],
    [
      51,
      30,
      0,
      39,
      0,
      "CONTROL_NET"
    ],
    [
      52,
      33,
      0,
      39,
      1,
      "IMAGE"
    ],
    [
      53,
      32,
      0,
      40,
      0,
      "CONTROL_NET"
    ],
    [
      54,
      34,
      0,
      40,
      1,
      "IMAGE"
    ],
    [
      55,
      41,
      0,
      35,
      1,
      "CONTROL_NET_STACK"
    ],
    [
      56,
      42,
      0,
      43,
      0,
      "MODEL"
    ],
    [
      57,
      42,
      1,
      43,
      1,
      "CONDITIONING"
    ],
    [
      58,
      42,
      2,
      43,
      2,
      "CONDITIONING"
    ],
    [
      59,
      42,
      3,
      43,
      3,
      "LATENT"
    ],
    [
      60,
      42,
      4,
      43,
      4,
      "VAE"
    ],
    [
      61,
      43,
      5,
      44,
      0,
      "IMAGE"
    ],
    [
      62,
      37,
      0,
      38,
      2,
      "CONTROL_NET_STACK"
    ],
    [
      63,
      38,
      0,
      39,
      2,
      "CONTROL_NET_STACK"
    ],
    [
      64,
      39,
      0,
      40,
      2,
      "CONTROL_NET_STACK"
    ],
    [
      65,
      40,
      0,
      42,
      1,
      "CONTROL_NET_STACK"
    ]
  ],
  "groups": [],
  "config": {},
  "extra": {
    "0246.VERSION": [
      0,
      0,
      4
    ]
  },
  "version": 0.4
}

测试图片
请添加图片描述

菜鸟也学大数据
关注
  • 7
    点赞
  • 踩
  • 4
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
专栏目录
[ComfyUI进阶教程] Controlnet预处理器节点安装和使用教程
细节控-AI先行者的博客
12-28 4526
虽然通过简单的提示词，在文生图和图生图中能生成不错的图片。但是如果要精准控图，实现人物姿势，深度图，画风迁移等进阶应用，就离不开ControlNet控图了。SD WEBUI实现精准控图，离不开ControlNet。高手进阶，AI绘画的灵魂。ControlNet的绘画模式是：先让用户输入一张参考图，AI根据指定的预处理模式，生成一张新图。常用 ControlNet预处理器包括：tile、openpos、depth、candy、lineart等。
[ComfyUI进阶教程] lcm+Lora+Controlnet工作流工作流
细节控-AI先行者的博客
01-20 1909
这是一个使用了LCM+lora加载器+CN（depth+tile）的工作流。#comfyui #ControlNet #工作流。配置了3个lora加载器，用来进行人物和风格设定。LCM lora加载器，加快生成图片的时间。网盘地址：专栏工作流--ControlNet。提示词编辑器，预制了默认的动态提示词。点击生成图片，组成了右侧的图片。1.tile用来进行风格控制。lcm+CN工作流.json。2.depth用来构图。
AI绘图的场景应用：文生图、图生图
qq_41314882的博客
03-15 1070
AIGC技术的未来发展前景广阔，随着人工智能技术的不断发展，AIGC技术也将不断提高。未来，AIGC技术将在游戏和计算领域得到更广泛的应用，使游戏和计算系统具有更高效、更智能、更灵活的特性。同时，AIGC技术也将与人工智能技术紧密结合，在更多的领域得到广泛应用，对程序员来说影响至关重要。未来，AIGC技术将继续得到提高，同时也将与人工智能技术紧密结合，在更多的领域得到广泛应用。感兴趣的小伙伴，赠送全套AIGC学习资料和安装工具，包含AI绘画、AI人工智能等前沿科技教程，模型插件，具体看下方。
ComfyUI 完全入门：ControlNet 使用教程
京东科技开发者
06-03 1186
大家好，我是每天分享AI应用的萤火君！今天继续给大家分享ComfyUI的入门必备技能：ControlNet。ControlNet 提供了十几种生成图片的控制方式，有的可以控制画面的结构，有的可以控制人物的姿势，还有的可以控制图片的画风，这对于提高AI绘画的质量特别有用；基于 ControlNet 的能力，炼丹师们可以将AI绘画拓展到很多的应用场景，比如艺术二维码、光影文字、线稿上色、老照片修复、图片风格转绘、IP图片创作等等。
Comfyui之ControlNet的使用
拥抱AIGC，拥抱Stable Diffusion，拥抱Comfyui工作流
07-09 943
Comfyui之ControlNet的使用
AI绘图:Controlnet在ComfyUI中的使用教程：提升Stable Diffusion可控性
04-01 3297
Controlnet是Stable Diffusion中一个至关重要的功能，它极大地增强了图像生成的可控性。通过引入额外的条件，Controlnet使我们能够更精确地控制生成的图像内容。本教程将详细介绍如何在ComfyUI中使用Controlnet，从而让你的创作过程更加灵活和精确。
【comfyui】40个工作流一次让你彻底掌握SD controlNet的使用场景(1)-线性处理器
bagell的博客
05-22 1547
Stable Diffusion 的 ControlNet 是一个控制预训练图像扩散模型的神经网络。它允许输入调节图像，然后使用该调节图像来操控图像生成。ControlNet 最近非常火，它和 Stable Diffusion 的结合使 Stable Diffusion 能够接受指导图像生成过程的条件输入，从而增强了 Stable Diffusion 的性能。使用 ControlNet 可以固定构图、定义姿势、描绘轮廓等，单凭线稿就能生成一张丰满精致的插画。
2024年SD ComfyUI零基础入门教程：controlNet工作流程全解析
最新发布
Yu91an的博客
07-12 810
本教程是 2024 年最新的SD ComfyUI入门指南，专为初学者设计，详细介绍了如何搭建controlNet工作流程。通过逐步指导，让您轻松掌握SD ComfyUI的高级功能，提升AI绘图的精准度和创造力。我们在WebUI中使用controlnet都是通过插件，而插件中包含了预处理器和模型。而在comfyui中我们是预处理和模型分开搭建的，接下来就让我们看看如何搭建吧首先我们先加载我们的默认工作流。
ComfyUI+AnimateDiff+ControlNet的Tile生成动画.zip
05-04
这个名为"ComfyUI+AnimateDiff+ControlNet的Tile生成动画.zip"的压缩包文件，显然涉及到几个关键概念，包括ComfyUI、AnimateDiff和ControlNet，以及Tile生成动画。我们将逐一探讨这些概念。 首先，ComfyUI可能是指...
mistoline和OpenPose及Depth的XL工作流
05-20
兄弟们记住这历史时刻 忘掉你手中的StableDiffusion1.5版本吧！ 只因StableDiffusion xl的controlnet终于迎来革命性的突破 ...然后我为大家准备了工作流 以及所有好用的控制模型也贴心奉上 亲爱的宝之门快去下载吧
ComfyUI+AnimateDiff+ControlNet的Openpose+Depth生成动画.zip
05-04
《基于ComfyUI、AnimateDiff、ControlNet与Openpose+Depth的动画生成技术详解》 在数字媒体领域，动画制作是一项复杂而精细的工作，需要融合艺术创作与技术应用。随着计算机视觉和人工智能技术的发展，自动化的动画...
AI-comfyui-工作流-绘画
04-11
内容：comfyui工作流，包含《4种局部重绘工作流》、《t2i_controlnet风格化》、《t2icontrolnet》、《test_cn_aux_full》、《局部保留重绘》、《标准i2i预览》、《标准t2i放大自动保存》、《标准t2i高清修复》、...
ComfyUI+AnimateDiff+ControlNet的Openpose生成动画.zip
05-04
《Openpose生成动画技术在ComfyUI、AnimateDiff与ControlNet中的应用》 在现代数字艺术和游戏开发中，动态角色的表现力对于提升用户体验至关重要。Openpose作为一种强大的人体关键点检测框架，为创建逼真的人物动画...
ComfyUI进阶篇：ControlNet核心节点
ai0604_2024的博客
07-10 1407
ComfyUI进阶篇：ControlNet核心节点前言：ControlNet_aux库包含大量的图片预处理节点，功能丰富，适用于图像分割、边缘检测、姿势检测、深度图处理等多种预处理方式。掌握这些节点的使用是利用ControlNet的关键，本篇文章将帮助您理解和学会使用这些节点。
全网最全comfyui工作流保姆级教程来啦！comfyui工作流搭建看这一篇就够了
m0_71746299的博客
05-10 9195
ControINet 有独立的控制图像，通过对图像的预处理，再结合提示词进行生成图像 不同的预处理可以对生成的图像进行不同的控制，一般有风格约束、线条约束、姿态约束、景深控制等，不管是WebUI还是ComfyUI，都有强大ControlNet，这里不详细介绍ControlNet，主要讲下如何在ComfyUI中使用ControlNet，后续可能专门写一篇ControlNet的详细介绍。当我们采用他人创建的工作流时，有时会遇到他们所使用的一些节点我们本地并未安装，从而出现了节点缺失的问题。
Controlnet在ComfyUI中的常见问题与解决方法
04-01 2945
熟练地在ComfyUI中使用Controlnet，为你的Stable Diffusion创作过程带来更高的精确度和灵活性。不断实践和探索，你将能够充分发挥Controlnet的潜力，本文介绍使用Controlnet的常见问题，现象和解决方法，供大家查阅。
【第8章】如何利用ControlNet生成“可控画面”？（配置要求/一键安装/快速上手/生成第一张图）ComfyUI基础入门教程
聚梦小课堂
06-12 454
在ComfyUI中使用ControlNet来生成可控画面，内容涵盖：ControlNet定义、用途、15大类别、安装等基础知识，并以一个“真人转二次元”的风格转绘工作流为例，带你一步步搭建一个，支持多个ControlNet混用的ComfyUI工作流...
探索3D姿态编辑的新纪元：ComfyUI 3D Pose Editor
gitblog_00099的博客
06-24 328
探索3D姿态编辑的新纪元：ComfyUI 3D Pose Editor 项目地址:https://gitcode.com/hinablue/ComfyUI_3dPoseEditor 在数字创意领域，每一个细微的创新都能引领全新的艺术表达方式。今天，我们要向您隆重介绍ComfyUI 3D Pose Editor——一个旨在扩展ComfyUI潜能的自定义节点工具集，让3D姿势编辑变得前所未有的直观与强...
comfyui controlnet inpaint
06-26
ComfyUI ControlNet Inpaint可能是指某种特定的软件或技术，但没有明确的信息表明它是什么。ControlNet是一种实时工业以太网协议，由Allen Bradley（现为Rockwell Automation）开发，主要用于设备之间的高速通信，特别是在自动化和控制系统的应用中。 Inpaint, 通常指的是图像修复或去噪技术中的“填补”或“恢复”部分，比如在摄影或图像处理中，当图像的一部分被遮挡或损坏时，使用算法来恢复这些区域的细节。在ComfyUI这样的用户界面库中，Inpaint技术可能会用于提供某种图像处理功能，帮助用户修复或美化显示的图像。 具体到"ComfyUI ControlNet Inpaint"，如果没有更多信息，这可能是ControlNet协议在ComfyUI框架下的某个高级功能，用于增强与工业设备通信时的数据可视化或处理能力。
评论
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值