AI Video Segment

Use a multimodal large model to understand video semantics and split shots from scene, shot-type, narrative, or subject perspectives. The API returns structured results by default. Set only_struct to false if you also want exported clip files.

Interactive Example

Create Task

Basic Information

Request Body

Mode Reference

Request Example

curl -X POST https://api.ai-mcn.tv:10000/task/video_ai_segment \
  -H "Authorization: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "file_id": "537489015178246",
    "mode": "scene",
    "min_duration": 1.5,
    "max_duration": 120,
    "only_struct": true
  }'

Success Example

{
  "code": 200,
  "msg": "success",
  "data": {
    "task_id": "537489015178247",
    "task_type": "video_ai_segment",
    "status": "queued"
  }
}

Query Task Result

Basic Information

Output Result (output_result)

shots Fields

files Fields

[
  {
    "id": 1,
    "file_id": "56817731376500742",
    "download_url": "https://api.ai-mcn.tv:9000/cloud/file/a6/56817731376500742.mp4"
  }
]

Success Example

{
  "code": 200,
  "msg": "success",
  "data": {
    "task_id": "537489015178247",
    "status": "completed",
    "progress": 100,
    "output_result": {
      "mode": "scene",
      "total_shots": 2,
      "total_duration": 18.6,
      "categories": [
        {
          "category": "Talking Head",
          "category_desc": "A person speaks directly to the camera",
          "shots": [
            {
              "id": 1,
              "subjects": ["Speaker", "Indoor background"],
              "tags": ["Talking", "Front-facing"],
              "shot_type": "Medium Shot",
              "desc": "A person speaks indoors",
              "start_time": 0.0,
              "end_time": 8.2,
              "confidence": 0.96,
              "thumbnail_time": 4.1,
              "category": "Talking Head",
              "category_desc": "A person speaks directly to the camera"
            }
          ]
        }
      ],
      "files": [
        {
          "id": 1,
          "file_id": "56817731376500742",
          "download_url": "https://api.ai-mcn.tv:9000/cloud/file/a6/56817731376500742.mp4"
        }
      ]
    },
    "create_time": "2026-04-09T08:00:00Z",
    "update_time": "2026-04-09T08:02:10Z"
  }
}

Error Codes

Usage Limits

• This API currently accepts a single video file per task
• Long videos are automatically split into overlapping chunks and merged on the server after analysis
• Structured output is returned by default. Set only_struct=false if you also need clip files