Skip to main content

Python bindings to the flandmark keypoint localization library

Project description

This package is a simple Boost.Python wrapper to the (rather quick) open-source facial landmark detector flandmark, version 1.0.7 (or the github state as of 10/february/2013). If you use this package, the author asks you to cite the following paper:

  author =      {U{\v{r}}i{\v{c}}{\'{a}}{\v{r}}, Michal and Franc, Vojt{\v{e}}ch and Hlav{\'{a}}{\v{c}}, V{\'{a}}clav},
  title =       {Detector of Facial Landmarks Learned by the Structured Output {SVM}},
  year =        {2012},
  pages =       {547-556},
  booktitle =   {VISAPP '12: Proceedings of the 7th International Conference on Computer Vision Theory and Applications},
  editor =      {Csurka, Gabriela and Braz, Jos{\'{e}}},
  publisher =   {SciTePress --- Science and Technology Publications},
  address =     {Portugal},
  volume =      {1},
  isbn =        {978-989-8565-03-7},
  book_pages =  {747},
  month =       {February},
  day =         {24-26},
  venue =       {Rome, Italy},
  keywords =    {Facial Landmark Detection, Structured Output Classification, Support Vector Machines, Deformable Part Models},
  prestige =    {important},
  authorship =  {50-40-10},
  status =      {published},
  project =     {FP7-ICT-247525 HUMAVIPS, PERG04-GA-2008-239455 SEMISOL, Czech Ministry of Education project 1M0567},
  www = {},

You should also cite Bob, as a core framework:

  author = {A. Anjos AND L. El Shafey AND R. Wallace AND M. G\"unther AND C. McCool AND S. Marcel},
  title = {Bob: a free signal processing and machine learning toolbox for researchers},
  year = {2012},
  month = oct,
  booktitle = {20th ACM Conference on Multimedia Systems (ACMMM), Nara, Japan},
  publisher = {ACM Press},
  url = {},


You can just add a dependence for xbob.flandmark on your to automatically download and have this package available at your satellite package. This works well if Bob is installed centrally at your machine.

Otherwise, you will need to tell buildout how to build the package locally and how to find Bob. For that, just add a custom egg recipe to your buildout that will fetch the package and compile it locally, setting the buildout variable prefixes to where Bob is installed (a build directory will work as well). For example:

parts = flandmark <other parts here...>
prefixes = /Users/andre/work/bob/build/debug


recipe = xbob.buildout:develop



To develop these bindings, you will need the open-source library Bob installed somewhere. At least version 1.1 of Bob is required. If you have compiled Bob yourself and installed it on a non-standard location, you will need to note down the path leading to the root of that installation.

Just type:

$ python
$ ./bin/buildout

If Bob is installed in a non-standard location, edit the file buildout.cfg to set the root to Bob’s local installation path. Remember to use the same python interpreter that was used to compile Bob, then execute the same steps as above.


Pretty simple, just do something like:

import bob
from xbob import flandmark

video ='myvideo.avi')
localizer = flandmark.Localizer()

for frame in video:
  print localizer(frame)

If you already have a detected bounding box, you can plug the coordinates of the bounding box into the localizer call:

landmarks = localizer(image, top, left, height, width)

In total, 8 landmarks are returned by the localizer. For the list and the interpretation of the landmarks, please have a look here.


Since version 1.1 of this package, the landmarks are returned in the Bob-typical order, which is (y,x). Please update your code to this new behavior.

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for xbob.flandmark, version 1.1.0
Filename, size File type Python version Upload date Hashes
Filename, size (2.1 MB) File type Source Python version None Upload date Hashes View

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring DigiCert DigiCert EV certificate Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page