The structure of the project is as following:
+-- data
| +-- iphone_ten
| +-- structure_sensor
+-- web
| +-- core_alg
| +-- out
| +-- test_picture.py
| +-- test_scan.py
+-- README.md
+-- .gitignore
The "./data" folder contains both 3D model from iPhone10 and structure sensor.
For example, to get an obj file taken from structure_sensor/picture/femur/. The director path is: data/structure_sensor/picture/femur
For example, to get an obj file taken from structure_sensor/scan/tibia/. The director path is: data/structure_sensor/scan/tibia
3. The IDE suggested to use is Pycharm 2020.1.2 x 64, and use pip install to install all packages in requirements.txt
The main file of measurement is in: msresearch/web/test_scan.py.
This file will take either single 3D model or multi 3D models as input, and output the measurements to a *.csv file.
Please specify global variables, such us bone type, input file name, and if the model taken from iPhone or structure sensor.
For example, to run measurement on femur_4.obj taken by sensor, set bone_type = Bone.Type.FEMUR, and index_default = 4, structure_sensor = True.
5. Take the output numbers and compare to ground truth from <msresearch/web/core_alg/utilities/results_anlysis.py>.
If the input are multi models, then it will also output a *.csv file which calculates mean values, absolute errors, standard derivations with ground truth
If the input is single model, error should be manually computed with a calculator