The project is a video processing utility. It identifies the object in the video from the list {SSD drive, Column, Sheet, Sheet with a number}, changes the color of this object to a negative and displays information about the time of its appearance in the report.
- ultralytics
- opencv-python
- torchvision
- numpy
The utility is designed to run on 🐧unix-like🐧 systems. If you have Windows, then you should install WSL (Windows Subsystem for Linux).
Tutorial:
- Windows features
- Turn on VMP (Virtual Machine Platform)
-
Make sure that you have enabled virtualization in your BIOS
-
Get Ubuntu WSL from Microsoft store
Before running the installation script, make sure that you have git installed, or install it using the command
sudo apt-get update
sudo apt-get install git -yInstallation script
sudo -s
git clone https://github.com/KirillMaltsev3341/AI_Video_Handler.git
cd AI_Video_Handler
./install_required_software.shInput files are located in input directory. Output files will be located in output directory. models directory contains models for detecting objects (objects_640.pt) and digits (digits_640.pt).
Initially, a file Видео для анализа.mov is already stored in the input directory (you can put your own file in the input directory for processing).
To check if the program is working:
./run.sh --input "Видео для анализа.mov"To find out more information about the launch flags:
./run.sh --help