{ "task_id": "d9zzvghifs95q8fkfd", "user_id": 1, "version": "fa9317a2ad086f7633f4f9b38f35c82495b6c5f38fa2afbe32d9d9df8620b389", "error": null, "total_time": 5, "predict_time": 5, "logs": null, "output": [ { "id": "ODaAdjDo02y", "status": "succeed", "type": "face_detect", "faces": [ { "id": 0, "link": "https://cdn.vmimgs.com/datarm/user/result/20250627/3c447d68-9668-4e31-9b54-1080a3dc12f9.jpg" }, { "id": 1, "link": "https://cdn.vmimgs.com/datarm/user/result/20250627/c5d7f7ef-0926-4c1e-b316-81df2313a2bf.jpg" } ], "started_at": 1751019524863, "finished_at": 1751019531542, "error": null } ], "status": "succeeded", "create_at": 1746492954, "completed_at": 1746493015, "input": { "source": "https://vmodel.ai/data/model/remaker/video-face-detect/tmp2ukv7myu.mp4" } }
Accurately detect faces in videos — frame-by-frame, automatically, and ready for post-processing.
The Video Face Detect API extends our powerful facial detection capabilities to video files. It automatically scans each frame of a video to detect one or more faces, and returns cropped face images extracted from the video over time. With the same interface as the Photo Face Detect API, it offers a drop-in experience for developers working with moving visual content.
Frame-by-frame detection
Analyzes each frame of the video to detect all visible faces.
Multi-face and multi-frame support
Detects multiple faces across multiple frames, with cropped images and tracking IDs.
Consistent API design
Shares the same input/output structure as the Photo Face Detect API — easy to migrate or integrate.
Cropped face image output
Returns a list of face image URLs per frame, pre-processed and ready for downstream tasks.
The Video Face Detect API accepts the same parameters as the Photo Face Detect API. Simply upload or reference a video file instead of an image.
detect_id
and an array of detected face image URLs