Video

Object Grouping

Segment and group objects across video frames

0K runs

Note: This playground provides simulated output to showcase functionality. No input data is processed or stored on our servers. Use this demo to explore the feature extractor's capabilities before integrating it into your application.

Input

File URL string

Enter a URL to a video file

Upload video

Drag and drop a video file here, or click to browse

Select File

# video_url string

Required

URL pointing to video file. Supported formats: MP4, MOV, AVI. Default: undefined

# min_object_size integer

Minimum object size in pixels to segment. Default: 50

# detection_threshold number

Confidence threshold for object detection (0.0-1.0). Default: 0.7

# tracking_threshold number

Similarity threshold for tracking objects across frames (0.0-1.0). Default: 0.8

# segmentation_mode string

Segmentation mode to use. Default: standard

# extract_attributes boolean

Extract object attributes (color, size, motion). Default: true

# generate_embedding boolean

Generate object embedding for similarity matching. Default: true

# crop_objects boolean

Generate cropped object images. Default: true

# temporal_consistency boolean

Enforce consistent segmentation across frames. Default: false

# frame_sampling integer

Process every Nth frame for performance. Default: 1

Output

{
  "document_id": "doc_abc123",
  "collection_id": "col_xyz789",
  "source_object_id": "obj_def456",
  "object": {
    "id": "obj_12345",
    "class": "person",
    "confidence": 0.92,
    "appearances": [
      {
        "frame_number": 120,
        "timestamp": 4,
        "mask": {
          "format": "rle",
          "data": "...",
          "width": 1920,
          "height": 1080
        },
        "bounding_box": {
          "x": 125,
          "y": 80,
          "width": 120,
          "height": 250
        }
      }
    ],
    "attributes": {
      "color": "red",
      "size_ratio": 0.15,
      "motion": "walking",
      "confidence": 0.85
    },
    "embedding": {
      "model": "object_encoder_v1",
      "dimension": 512,
      "vector": [
        0.12,
        0.34
      ],
      "normalized": true
    },
    "representative_image_url": "https://storage.example.com/objects/obj_12345.jpg"
  },
  "video": {
    "filename": "street_scene.mp4",
    "width": 1920,
    "height": 1080,
    "fps": 30,
    "duration": 180.5
  }
}

Already have embeddings? Skip extraction — search your own vectors with MVS. First 1M vectors free.

Try MVS →