NetraFlow
Capabilities

Objects

Detect objects in video with categories, confidence scores, and proof frames.

Object detection identifies items appearing in video frames — electronics, furniture, vehicles, clothing, and more. Video frames are analyzed to return categorized objects with confidence scores and timestamps.

curl -X POST https://api.netraflow.com/v1/jobs \
  -H "Content-Type: application/json" \
  -H "X-Api-Key: sk_live_your_key_here" \
  -d '{
    "url": "https://www.youtube.com/watch?v=example",
    "capabilities": ["objects"]
  }'
{
  "data": {
    "job_id": "job_abc123",
    "status": "completed",
    "results": {
      "objects": [
        {
          "name": "laptop",
          "category": "electronics",
          "confidence": 0.92,
          "total_appearances": 5,
          "appearances": [
            {
              "timestamp": 15.0,
              "context": "Silver MacBook on desk, screen visible"
            },
            {
              "timestamp": 45.0,
              "context": "Presenter typing on laptop during demo"
            }
          ]
        },
        {
          "name": "whiteboard",
          "category": "office",
          "confidence": 0.85,
          "total_appearances": 3,
          "appearances": [
            {
              "timestamp": 30.0,
              "context": "Whiteboard with diagrams in background"
            }
          ]
        }
      ]
    }
  }
}

Object fields

Prop

Type

Appearance fields

Prop

Type

Proof frames

Set return_proof: true to receive proof frame URLs for high-confidence detections. The same rules apply as for brand proof frames: confidence >= 0.8, secondary AI verification, up to 20 frames per job.

curl -X POST https://api.netraflow.com/v1/jobs \
  -H "Content-Type: application/json" \
  -H "X-Api-Key: sk_live_your_key_here" \
  -d '{
    "url": "https://www.youtube.com/watch?v=example",
    "capabilities": ["objects"],
    "return_proof": true
  }'
{
  "data": {
    "job_id": "job_abc123",
    "status": "completed",
    "results": {
      "objects": [
        {
          "name": "laptop",
          "category": "electronics",
          "confidence": 0.92,
          "total_appearances": 2,
          "appearances": [
            {
              "timestamp": 15.0,
              "frame_url": "/v1/frames/job_abc123/object_laptop_15.0.jpg",
              "bounding_box": { "x": 0.25, "y": 0.4, "width": 0.3, "height": 0.2 },
              "context": "Silver MacBook on desk"
            },
            {
              "timestamp": 45.0,
              "frame_url": "/v1/frames/job_abc123/object_laptop_45.0.jpg",
              "context": "Presenter typing on laptop"
            }
          ]
        }
      ]
    }
  }
}

Proof frame URLs are authenticated — fetch them with the same X-Api-Key header.

On this page