About
DaVid is a simple cross-platform GUI for annotating robotic and endoscopic surgical actions for use in deep-learning research.
Features
- Simple and lightweight interface for loading and annotating surgical frames for Windows, Linux and macOS.
- Joystick controls for intuitive and rapid action annotation.
- Keyboard shortcuts support.
- Hassle-free data export.
Installation
In order to run DaVid, you'll first have to make sure you're running Python 3.5 or above.
python -V
After downloading and unzipping the contents of this repository, navigate to the destination folder and install dependencies using:
cd DaVid-main
pip install –upgrade pip
pip install -r requirements.txt
You can now launch DaVid with:
python DaVid.py
Download (Coming soon)
Alternatively, you can download pre-compiled versions of the application for macOS and Windows from here.
Usage
- After launching DaVid, select a video file and wait while frames are extracted and saved. Note that this process only happens once per video and should take up to a few minutes.
- For the current frame move the controls accordingly.
- That's it! Move on to the next frame.