Skip to content

  • Projects
  • Groups
  • Snippets
  • Help
  • This project
    • Loading...
  • Sign in / Register
deformetrica
deformetrica
  • Overview
    • Overview
    • Details
    • Activity
    • Cycle Analytics
  • Repository
    • Repository
    • Files
    • Commits
    • Branches
    • Tags
    • Contributors
    • Graph
    • Compare
    • Charts
  • Issues 11
    • Issues 11
    • List
    • Board
    • Labels
    • Milestones
  • Merge Requests 0
    • Merge Requests 0
  • CI / CD
    • CI / CD
    • Pipelines
    • Jobs
    • Schedules
    • Charts
  • Wiki
    • Wiki
  • Snippets
    • Snippets
  • Members
    • Members
  • Collapse sidebar
  • Activity
  • Graph
  • Charts
  • Create a new issue
  • Jobs
  • Commits
  • Issue Boards
  • aramislab
  • deformetricadeformetrica
  • Wiki
  • Home

Home

Last edited by Alexandre Sep 16, 2018
Page history

Introduction

[NEW!] A graphical user interface (GUI) is now available, in its alpha version! Launch it with the deformetrica gui command. We warmly welcome any feedback about this new feature.

The Deformetrica applications are called with command lines of the form:

  • deformetrica estimate model.xml data_set.xml [-p optimization_parameters.xml] or
  • deformetrica compute model.xml [-p optimization_parameters.xml]

where:

  • deformetrica is an alias for python deformetrica.py: the main script of the software, in the src folder, automatically created during the conda installation.
  • model.xml is an xml file containing the description of the task that is to be performed (such as estimating a registration, an atlas, ...) as well as some hyper-parameters for the objects and the deformations used.
  • data_set.xml is an xml file containing the paths to the input objects from which a statistical model will be estimated. The shooting and parallel transport applications are direct computations, and do not need this xml file.
  • optimization_parameters.xml contains optional details about the optimization method.

Additional arguments in the command line are:

  • --output=path/to/output/folder used to specify the output directory i.e. where all results are saved.
  • -v DEBUG: allow to get more printed information about the current computation.

Tutorials and examples are available to demonstrate how the three xml files should be filled. Exhaustive lists of options are given in the user manual.

The Deformetrica applications theoretically rely on a control-points-based instance of the Large Deformation Diffeomorphic Metric Mapping (LDDMM) framework. Find more details on the dedicated wiki page.

Contents

  • The LDDMM framework: a description of the mathematics behind Deformetrica.
  • Tutorials and examples for each functionality:
    • Registration
    • Deterministic atlas
    • Bayesian atlas
    • Geodesic regression
    • Parallel transport
    • Shooting
    • Longitudinal atlas [work in progress]
  • User manual:
    • Available object types: a list of all available shapes in Deformetrica and how to use them
    • List of options for the model.xml file
    • List of options for the data_set.xml file
    • List of options for the optimization_parameters.xml file
    • Estimation methods
    • About the kernels
Clone repository
  • 1_lddmm
  • Home
  • 2_tutorials
    • 2.1_registration
    • 2.2_deterministic_atlas
    • 2.3_bayesian_atlas
    • 2.4_regression
    • 2.5_parallel_transport
    • 2.6_shooting
    • 2.7_longitudinal_atlas
  • 3_user_manual
    • 3.1_available_objects
    • 3.2_model_xml_file
    • 3.3_data_set_xml_file
    • 3.4_optimization_parameters_xml_file
    • 3.5_estimators
    • 3.6_kernels
More Pages
×

New Wiki Page

Tip: You can specify the full path for the new file. We will automatically create any missing directories.