How to make ANY PTZ VISCA camera to autotrack human speaker on boring conferences

This script works with any (i hope) PTZ cameras with VISCA protocol inside.

1. Find your camera IP address
2. Find the syntax of VISCA commands in your camera's manual
3. Start RTSP feed from your camera
4. Print in all attributes into the code
5. Start the script
6. Enjoy
7. Also please install ffmpeg

Video EN
Video RU

License information

This code is developed by Sergey Korneyev from AVStream (https://avstream.ru). It is distributed under an open-source license, allowing for free use, modification, and distribution, provided that this notice and attribution to the original author are preserved.

This code utilizes FFmpeg, an open-source multimedia framework licensed under the LGPL or GPL. FFmpeg is a trademark of Fabrice Bellard and is maintained by the FFmpeg team. For more details on FFmpeg's licensing terms, please refer to https://ffmpeg.org/legal.html.

This code also uses YOLOv8 by Ultralytics, an open-source object detection model. YOLOv8 and Ultralytics' models are distributed under the GPL-3.0 License. For more details on YOLOv8's licensing, please refer to https://github.com/ultralytics/ultralytics.

By using this code, you acknowledge that you comply with the terms and conditions of both FFmpeg's and YOLOv8's licenses, as well as the open-source license provided by Sergey Korneyev. Redistribution of this script with modifications should maintain these notices to respect the original work and licenses of the authors.

Author: Sergey Korneyev, AVStream
Website: https://avstream.ru
Telegram: https://t.me/Kvanterbreher