Metadata Parser

gst-ai-metadata-parser-example performs object detection and parses the detection metadata (coordinates, class, confidence) into structured output, suitable for integration with external analytics systems or logging.

Prerequisites

• Completed QIM SDK Installation and Model Download

Steps

1. Verify Model

ls -l /etc/models/yolox_quantized.tflite

2. View Configuration

cat /etc/configs/config-metadata-parser.json

3. Run

radxa@airbox$

gst-ai-metadata-parser-example --config-file=/etc/configs/config-metadata-parser.json

Press Ctrl + C to stop.

Expected Output

Running app with model: /etc/models/yolox_quantized.tflite and labels: /etc/labels/yolox.json
Using DSP delegate
VERBOSE: Replacing 329 out of 329 node(s) with delegate (TfLiteQnnDelegate) node
Pipeline state changed from PAUSED to PLAYING

The display shows the detection view while the terminal streams structured detection metadata (JSON format).

Validation

• YOLOX model (329 ops) running on DSP
• Pipeline reaches PLAYING state
• Terminal outputs valid metadata structures