Create several defaced variants of images

[hjmjohnson]

User Story

As a data curator, I want to deface images with various algorithms so that we can build robust algorithms for defaced data.

Detailed Description

Tasks to be completed

• [x] Deface with a variant of "https://github.com/BRAINSia/BIDSonym" "run.py" that does not require bids data structure. create "pydeface", "mri_deface", "deface_wf", "mridefacer" variants of each image.
• [x] Deface with https://nrg.wustl.edu/software/face-masking/ (https://hub.docker.com/r/mohanar/facemasking/tags)
• [ ] Train for landmark finding (including eye finding) even when eyes are removed by defacer.

Acceptance Criteria

• [x] A complete set of defaced images are created

[ArjitJ]

https://github.com/ArjitJ/BIDSonym

The files are in files.txt, the run script takes in one argument(path of the image)

[ArjitJ]

The nrg defacing outputs are not properly aligned with the inputs. (Do not overlap in slicer). Need to perform registration before using in the dataset