Automatic Alignment of Large Collections of Anatomical Surfaces for Geometric Morphometrics
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
code
software
viewer
.gitignore
README.md

README.md

Puente Alignment

Orignally written for the MST-based Generalized Dataset Procrustes Distance by Jesús Puente; this current version contains experimental features involving two synchronization-based alignment algorithms (spectral/SDP relaxation).

MATLAB Code originally written by Jesús Puente (jparrubarrena@gmail.com); currently maintained by Tingran Gao (trgao10@math.duke.edu) and Julie Winchester (julia.m.winchester@gmail.com). This code has also been ported to R by Christopher Glynn (glynn@stat.duke.edu) under the name auto3dgm.


Sequential Execution

The entry point is the script code/main.m; see comments at the top of that script for a quick introduction.


Parallel Execution

The current version of PuenteAlignment supports parallel computations on a cluster managed by Sun Grid Engine (SGE). To enable parallel execution, follow the steps 1 to 6 below.

  1. Get the current version of PuenteAlignment. Simply cd into your desired path, then type

     git clone https://github.com/trgao10/PuenteAlignment/
    
  2. Find script jadd_path.m in the folder PuenteAlignment/code/, and set paths and parameters there. If you assign an email address to the varialbe email_notification, a notification will be sent automatically to that email address whenever a cluster job completes or aborts.

  3. Launch MATLAB, cd into the folder PuenteAlignment/code/, type in clusterPreprocess and press ENTER. All jobs should then be submitted to the cluster. Use qstat to monitor job status.

  4. After all jobs are completed, type in clusterMapLowRes and press ENTER.

  5. After all jobs are completed, type in clusterReduceLowRes and press ENTER. This generates low-resolution alignment results in the output folder you specified in jadd_path.m.

  6. Type in clusterMapHighRes and press ENTER to submit high-resolution alignment jobs to the cluster. Use qstat to monitor job status.

  7. After all jobs are completed, type in clusterReduceHighRes and press ENTER. This generates high-resolution alignment results in the output folder you specified in jadd_path.m.


WebGL-based Alignment Visualization

After the alignment process is completed, the result can be visualized using a javascript-based viewer located under the folder viewer/. See here for an online demo.

  1. Move all output files ending with "_aligned.obj" from the subfolder aligned/ (under your output folder) to the subfolder viewer/aligned_meshes/.
  2. Set up an HTTP server under the folder viewer/. (If you already placed the folder viewer/ somewhere with HTTP services, feel free to skip this step.) For instance, you can cd viewer/ and type into the terminal python -m SimpleHTTPServer 8000 if you are using Python 2.x, or equivalently python -m http.server 8000 if you have Python 3.x.
  3. Launch your browser and direct it to http://localhost:8000/auto3dgm.html.

Mosek License File

You will need a mosek license for using the fast linear programming routine for pairwise alignments. If you have an academic/institutional email address, you are eligible for a free academic license from mosek.com. Upon receiving the mosek license, simply drop it under the folder PuenteAlignment/software/mosek/.


Please Cite:

Boyer, Doug M., et al. A New Fully Automated Approach for Aligning and Comparing Shapes. The Anatomical Record 298.1 (2015): 249-276.

Puente, Jesús. Distances and Algorithms to Compare Sets of Shapes for Automated Biological Morphometrics. PhD Thesis, Princeton University, 2013.