Examples¶

These examples begin with a working application and then explain the relevant API. Start with the quickstart if you have not run the generated demo yet.

Inspect and debug a recording¶

Open the tutorial

Run the verifier against a public FMV sample, interpret its findings, open the same stream in the player, and export its metadata.

Verifier report with stream inventory and findings

Create FMV from video and metadata¶

Open the tutorial

Combine an ordinary video and ArcGIS-style metadata CSV into an MPEG transport stream, then verify and play the result.

Add AI detections as VMTI¶

Open the tutorial

Run multiple detector stages, attach target identity and model information, and encode the result as ST 0903 VMTI.

Player with VMTI detections from an Ultralytics model

Build a video, map, and activity dashboard¶

Open the tutorial

Prepare browser media and a synchronized metadata timeline, then render video, telemetry, map features, and activity in a small web application.

FMV dashboard with video, telemetry, map, and activity

Transform and fan out live KLV¶

Open the tutorial

Process arbitrarily chunked transport data, write the transformed stream, and publish the same timed metadata to a UI or GIS consumer.

More runnable examples¶

The source distribution and GitHub repository contain focused examples for:

PyAV video and audio decoding;
Ultralytics, ONNX Runtime, Triton, and JSON/HTTP inference;
frame-to-KLV correlation and VMTI emission;
the standalone HTML dashboard.

Clone the repository only when you want to run or modify those source examples. Installing the library itself requires only PyPI.