Skip to content

Installation

This guide provides instructions to install ViVa-DataCreator.

System Requirements

The code automatically creates a Python 3.12 virtual environment using uv, even if a different Python version is installed on your system.

1. Setting UV, the Python project manager

To facilitate the creation of virtual environments and manage Python packages and their dependencies we use a state of the art framework uv, its installation is straightforward and can be done via the following command:

Using curl 

curl -LsSf https://astral.sh/uv/install.sh | sh
Using wget 
wget -qO- https://astral.sh/uv/install.sh | sh

Use irm to download the script and execute it with iex: 

powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

2. Install ViVa-DataCreator

Choose one of the following installation methods:

Recommended for most users

Install the latest stable release from the Python Package Index (PyPI).

uv venv --python 3.12
uv pip install viva-datacreator --upgrade

Recommended for developers

Install the latest development version directly from the GitHub repository.

git clone https://github.com/viva-safeland/viva_datacreator.git
cd viva_datacreator
uv sync

Automatic Setup

The application will automatically download the required SAM2 checkpoints on first launch if they are not already available. No manual download is required.

4. Video Requirements

To use ViVa-DataCreator, you need video files for processing. The application accepts various video formats. Below the recommended specifications are listed:

Video specifications:

  • Format: MP4, AVI, MOV, MKV, WMV, MPEG, MPG, FLV, WebM, and other formats supported by the OpenCV library
  • Resolution: 1080p or higher recommended for better segmentation quality
  • Content: Videos containing objects you want to segment (people, vehicles, animals, etc.)