Hololens 2 labeling base

Overview

These tools are used to receive and save data from the Hololens 2, the working application can be found under

https://github.com/intuitive-robots/human-demonstration-ar/tree/labeling

Structure

  ___
    |- calibration         # after a camera calibration the result will be stored here 
    |- calibration-frames  # Only the images put here will be used by calibration.py
    |- doc                 # documentation 
    |- imgs                # single frames from video2label.py will be safed here 
    |- saves               # Uploaded data from the hololens will be put here
    |- videos              # videos to test can be stored here

Scripts

calibration.py

• Extracts camera calibration data from the images saved in calibration-frames
• Images have to contain an calibration checkerboard (https://markhedleyjones.com/projects/calibration-checkerboard-collection)
• Outputs the data to calibration\calibration.yaml

drifter.py

• Opens a websocket server to receive hololens connections
• Saves incoming data to a file in saves/ (the last saved file will be stored in saves/meta.yaml)

fileio.py

• Utility for file loading and saving

video2label.py

• Visualizer for the saved data