Forgot password? Request a new one here!

Page navigation

Welcome to the femur project page of the FutureLearn course on Statistical Shape Modelling.

Here you can find all the details on how to download the data required at the different steps of the project, and on how to participate in the femur reconstruction competition.

Downloading data (step 2 of the project)

Here, you can download the dataset of 50 misaligned femur triangle meshes and their corresponding landmarks.

You need to log in to download the data!

Downloading data (step 3 of the project)

Here you can download the dataset of 10 partial femur shapes to be completed using your previously built femur model.

You need to log in to download the data!

Participating in the competition:

Once you have completed the reconstruction for all 10 shapes, you can take part in this competition and compare the quality of your reconstructions to those of your fellow learners.


  • All participants obtain a set of 50 training examples, as well as a set of 10 triangle meshes of partial femur shapes

  • The goal of the competition is to obtain the best possible reconstruction (shape completion) for the 10 given partial femurs

  • Participants are required to hand in the reconstructed 10 shapes.

    • Make sure that each completed mesh is rigidly aligned with its original incomplete mesh.

    • Make sure that your meshes are in correspondence with the reference mesh of the femur model computed at step 1 of the project

  • The evaluation is performed once the reconstruction is uploaded on the Sicas Medical Image Repository (SMIR)

  • All 10 reconstructions need to be uploaded to the SMIR, in order for a participation to be valid.


All of the provided 10 partial shapes were obtained by performing mesh clipping on complete femur shapes that we refer to as ground truths. The quality of the reconstruction is therefore evaluated by comparing the reconstruction to the ground truth according to 2 mesh distance measures described below.

Important: these distances will be measured only for points belonging to the patches that need to be reconstructed, and not on the entire shape

1. Average mesh distance to ground truth

For every point of the first mesh, its distance to the closest point on the second triangle mesh is computed. The average over all points is returned. The average distance will be evaluated symmetrically, i.e. from the ground truth to the reconstruction, and from the reconstruction to the ground truth. The returned value is the average of the evaluation in both directions.


The lower the value, the better the reconstruction.

2. Hausdorff mesh distance to ground truth

The Hausdorff distance is computed between the reconstruction and the corresponding ground truth.

Both metrics are implemented as part of the Scalismo library. The source code is available here.


For each submitted reconstruction, both the Hausdorff and the average mesh metrics are evaluated.

Ranking of the participant is done according to their performance (e.g. if there are 10 participants, the best performing algorithm gets 1 point and the worst performing algorithm gets 10 point). Each participant is therefore ranked 20 times. The final ranking will be the average of the individual rankings.


Reconstruction format:

Reconstructions are to be submitted as triangle meshes, in the STL file format

Reconstrution naming:

1. You have to name your results exactly like this template NAMESPACE.description.vsdid.stl: For case 1 use VSD.case_1.101147.stl. Here's the list for all cases:


2. Upload your files using the Upload & Evaluation page

3. Multiple versions: To keep track ouf your versions, you can change the description part in the filename.

A 1st version:


and a modified 2nd version:


4. Additional information for the template definition

	value: VSD

	user defined
	no special character except undersocores `_` or dashes `-`
	example: case_1

vsdid (of the partials image)
	static per case
	example: 101147

	value: .stl

The code used for the evaluation can be found here.

Evaluation Results