ARACAM

A RGB-D Multi-View Photogrammetry System for Lower Limb 3D Reconstruction Applications

https://mdpi-res.com/d_attachment/sensors/sensors-22-02443/article_deploy/sensors-22-02443-v2.pdf

Dependencies:

ARACAM uses different libraries, install dependencies using PyPi.

pip install nbimporter
pip install opencv-contrib-python
pip install matplotlib
pip install scikit-image
pip install open3d
pip install pandas
pip install pyvista
pip install pymeshfix
pip install pymeshlab
pip install plotly
pip install kaleido

Pipeline

The main.ipynb notebook contains the pipeline of the project and it is structured as follows:

1. segmentation_color.ipynb - Selects the desired object by means of depth and color.

2. to_pointcloud.ipynb - Transforms all color and depth pair information into 3D point cloudls.

3. pointcloud_process.ipynb - Reconstructs all poinclouds using ICP algorithm

4. 3D_validation.ipynb - Compares the original mesh obtained from a CT vs the ARACAM

Model comparisson

Original model obtained from a CT image

Model obtained from the ARACAM system

Hausdorff distances was used to compare both models.

Citing

If you use the ARACAM work for a scientific purpose, please cite the following paper.

@Article{Barreto2022,
	AUTHOR = {Barreto, Marco A. and Perez-Gonzalez, Jorge and Herr, Hugh M. and Huegel, Joel C.},
	TITLE = {ARACAM: A RGB-D Multi-View Photogrammetry System for Lower Limb 3D Reconstruction Applications},
	JOURNAL = {Sensors},
	VOLUME = {22},
	YEAR = {2022},
	NUMBER = {7},
	ARTICLE-NUMBER = {2443},
	URL = {https://www.mdpi.com/1424-8220/22/7/2443},
	PubMedID = {35408058},
	ISSN = {1424-8220},
	DOI = {10.3390/s22072443}
}