Data

Data Organization

Data Processing

Essential files to know in this directory are listed below. Instructions on running individual scripts are found in the Python files. Its is recommened a Python 3.8.10 virtual environment named env is created at the base of the Data directory.

1. ProgressTracker.py - Python application used to manually track sub-task completion time. Will need to be updated to match task structure in Unity project or own custom tasks.
2. RedrawWallArt.py - Python application used to show participants drawing process.
3. RetrofittingParser.py - Single-subject data processing. Turns raw HoloLens 2 Unity output into a more usable .csv file. Should be run on every subject.
4. RetrofittingPlotter.py - Debug plotting to quickly view data. Questionable validity. Mostly carried over from another project and havent tested in a while.
5. WallArtTool.py - Python Application used to draw, and drawings can be later re-drawn using another script. Designed for a Microsoft Surface Pro 4 and pens that work on that system. Cannot guarantee it works on other systems.
6. WallEstimator.py - Python application where estimated wall cutout ares are made by participants.

Batch files used to quickly some of these applications are included as well.

1. RunDrawingApp.bat - Launches drawing application script.
2. RunProgressTracker_AR.bat - Launches progress tracker script.
3. RunProgressTracker_Paper.bat - Launches progress tracker script.
4. RunScripts.bat - Launches single participant data processing and plotting scripts.
5. RunWallEstimatorApp.bat - Launches wall estimator script.

Recommended Structure

Subject data should be placed in folders in the Subjects folder. Every subject has their own folder titled after their integer Participant ID. An example of what a typical subject directory should look like is shown in the image below.

1. Photos - Contains the pictures of the wall used to record participant progress.
2. 1.pdf - Consent form.
3. 1_2022-11-21_10-59-18-021195.txt - Drawing application output data.
4. 11-21-2022_10-31-51_AUTO.csv - Python-processed experiment data.
5. 11-21-2022_10-31-51_AUTO.txt - Raw experiment data from HoloLens 2.
6. 20230110_135717.jpg - Image of notes taken during study.
7. ProgressTracker_MANUAL_11_21_2022-10_32_31_518410.csv - Output from experiment progress tracker Python application.
8. screenshot-2022-11-21_11-01-23-2460820001.jpg - Image of drawing application drawing.
9. SizeEstimation_1_POST.csv - Post-experiment cutout area estimation (Python application output).
10. SizeEstimation_1_PRE.csv - PRE-experiment cutout area estimation (Python application output).

Camera Info

Images captured by a camera were used for wall cutout area estimation. The images are first perspective/calibration corrected using OpenCV and then the cutout areas were manually tagged using the PerspectiveCorrection.py script.

Special thanks to the following sources:

1. https://pyimagesearch.com/2014/08/25/4-point-opencv-getperspective-transform-example/
2. https://pysource.com/2021/05/28/measure-size-of-an-object-with-opencv-aruco-marker-and-python/#

Participant Plotting

Transform and study conditions are logged by the HoloLens 2 at a rate of 5 hz. Once this data is processed and converted to a CSV file (see above), it can be plotted in Rhinocerous 6 and Grasshopper. These script and a primitive re-creation of the test environment are found in the RhinoPlotting folder.