This ComfyUI workflow is designed to convert video footage into a temporally stable pose control map using the OpenPose technique. It leverages the DWPose model to accurately detect and map human poses throughout the video sequence. The workflow utilizes several key nodes, including DWPreprocessor for initial data preparation, VHS_LoadVideo for importing video files, and VHS_VideoCombine for merging processed frames into a cohesive output. The use of comfyui-videohelpersuite and comfyui_controlnet_aux custom nodes ensures enhanced video processing capabilities and pose detection accuracy. This workflow is particularly useful for applications requiring detailed human pose analysis, such as animation, video editing, and motion capture.
API
從程式碼使用此工作流
每個 Comfy 工作流都是一個 JSON 圖。下方的 payload 就是此工作流，與 ComfyUI 執行時完全相同 — 你可以從此 URL 取得、放入版本控制，或在 ComfyUI 載入並逐節點執行。
使用 Comfy SDK 以 TypeScript 或 Python 執行。相同程式碼可用於 Comfy Cloud 或你自架的 ComfyUI — 只需更改 base URL。
// Install (beta)
npm i @comfyorg/sdk
// Run "影片到姿態對映 - OpenPose" (TypeScript)
import { Comfy } from "@comfyorg/sdk";
const client = new Comfy({ apiKey: "comfyui-..." });
// This workflow, exported in API format (see note below)
const wf = await client.workflows.fromFile("utility-openpose-video_api.json");
const job = await client.run(wf);
await job.getOutputs("4")[0].toFile("output.png"); // VHS_VideoCombineSDK 需使用 API 格式的工作流：請在 ComfyUI 開啟此工作流，並使用 檔案 → 匯出工作流（API）。
常見問題
