• 版本 [version]
• 下载 78
• 文件大小 35086
• 文件计数 1
• 创建日期 2026年6月9日
• 最后更新 2026年9月21日

油画转绘工作流workflow

这是一份开箱即用的万物迁移工作流（SDXL、IPAdapter、ControlNet），拖入 ComfyUI 即可运行，全图 31 节点 / 51 连线。核心流程 PreviewImage ×4、ControlNetApply ×2、CLIPTextEncode ×2，输出图像，参数已调好，替换输入图即可复现效果。主模型 depth_anything_vitl14.pth、sam_vit_b_01ec64.pth、bbox/face_yolov8m.pt。已挂载 LoRA：sdxl油画/oil_painting.safetensors、sdxl油画/印象派EldritchImpressionismXL1.5.safetensors。下载后直接导入即可使用，建议显存 8GB 起。

• 主模型：depth_anything_vitl14.pth、sam_vit_b_01ec64.pth、bbox/face_yolov8m.pt
• LoRA / 适配器：sdxl油画/oil_painting.safetensors、sdxl油画/印象派EldritchImpressionismXL1.5.safetensors
• 核心节点：PreviewImage ×4、ControlNetApply ×2、CLIPTextEncode ×2
• 输出形式：输出图像
• 规模：31 节点 / 51 连线
• 建议显存：8GB 起
• 适用场景：风格迁移、IPAdapter 参考图迁移、老照片修复与万物转绘。

工作流节点图

图较大可横向滚动；点击图片查看高清矢量原图（浏览器中可自由缩放）。

资源头图

使用说明

下载工作流文件后在 ComfyUI 中直接拖入画布（或菜单 Load 导入）即可运行；缺失的自定义节点用 ComfyUI Manager 一键安装，模型按文件名放入对应的 models 目录。

源码预览

{
    "last_node_id": 166,
    "last_link_id": 278,
    "nodes": [
        {
            "id": 40,
            "type": "ControlNetApply",
            "pos": [
                540,
                140
            ],
            "size": [
                317.4000244140625,
                98
            ],
            "flags": [],
            "order": 28,
            "mode": 0,
            "inputs": [
                {
                    "name": "conditioning",
                    "type": "CONDITIONING",
                    "link": 146,
                    "label": "条件"
                },
                {
                    "name": "control_net",
                    "type": "CONTROL_NET",
                    "link": 56,
                    "slot_index": 1,
                    "label": "ControlNet"
                },
                {
                    "name": "image",
                    "type": "IMAGE",
                    "link": 59,
                    "label": "图像"
                }
            ],
            "outputs": [
                {
                    "name": "CONDITIONING",
                    "type": "CONDITIONING",
                    "links": [
                        221
                    ],
                    "shape": 3,
                    "slot_index": 0,
                    "label": "条件"
                }
            ],
            "properties": {
                "Node name for S&R": "ControlNetApply"
            },
            "widgets_values": [
                0.8
            ]
        },
        {
            "id": 90,
            "type": "CannyEdgePreprocessor",
            "pos": [
                210,
                270
            ],
            "size": [
                315,
                106
            ],
            "flags": [],
            "order": 12,
            "mode": 0,
            "inputs": [
                {
                    "name": "image",
                    "type": "IMAGE",
                    "link": 150,
                    "label": "图像"
                }
            ],
            "outputs": [
                {
                    "name": "IMAGE",
                    "type": "IMAGE",
                    "links": [
                        149,
                        151
                    ],
                    "shape": 3,
                    "label": "图像",
                    "slot_index": 0
                }
            ],
            "properties": {
                "Node name for S&R": "CannyEdgePreprocessor"
            },
            "widgets_values": [
                100,
                200,
                512
            ]
        },
        {
            "id": 38,
            "type": "DepthAnythingPreprocessor",
            "pos": [
                540,
                280
            ],
            "size": [
                314.3816833496094,
                82.00997161865234
            ],
            "flags": [],
            "order": 11,
            "mode": 0,
            "inputs": [
                {
                    "name": "image",
                    "type": "IMAGE",
                    "link": 145,
                    "label": "图像"
                }
            ],
            "outputs": [
                {
                    "name": "IMAGE",
                    "type": "IMAGE",
                    "links": [
                        55,
                        59
                    ],
                    "shape": 3,
                    "slot_index": 0,
                    "label": "图像"
                }
            ],
            "properties": {
                "Node name for S&R": "DepthAnythingPreprocessor"
            },
            "widgets_values": [
                "depth_anything_vitl14.pth",
                1024
            ]
        },
  
… (完整内容请下载文件查看)

代码较长，以上仅前 3800 字符 —— 点击下方按钮下载完整 JSON。