GStreamer Object Detector

A window displaying a still of a video of several people walking in a crowded hall. The people in the foreground are surrounded by orange boxes labeled "Person".

These samples use the Xnor C and Python bindings with GStreamer to display a window with live-updating object detection results. They draw colorful bounding boxes in real time around people, animals, or things in the scene (depending on which Xnor model you use). To run these samples, you must have GStreamer installed and a webcam or video file available on your device.

Language

C, Python

Location

xnor-sdk-linux-x86_64/samples/c/gstreamer_live_overlay_object_detector.c, xnor-sdk-linux-x86_64/samples/python/gstreamer_live_overlay_object_detector.py

Dependencies

pycairo (Python only), pygobject (Python only), GStreamer (via system package manager), an installed Xnor detector model.

Compatible Xnor Models

All object detection models

Example Usage

python3 gstreamer_live_overlay_object_detector.py --video_file ~/Videos/dog.mp4