An Automatic Hybrid Segmentation Approach for Aligned Face Portrait Images

Martin Hirzer, Martin Urschler, Horst Bischof, Josef Alois Birchbauer

Research output: Chapter in Book/Report/Conference proceedingConference paperpeer-review


With the introduction of electronic personal documents (e.g. passports) in recent years the analysis of suitable photographs has become an important field of research. Such photographs for machine readable travel documents have to fulfill a set of minimal quality requirements defined by the International Civil Aviation Organization (ICAO). As some of the specified requirements are related to certain image regions only, these regions must be located in advance. In this work we present an automatic segmentation method for aligned color face images. The method is based on a convex variational energy formulation which is solved using weighted total variation. We apply constraints in the form of prior knowledge about the spatial configuration of typical passport photographs in order to solve for a global energy minimum. Several experiments on face images from two different datasets are presented to evaluate the performance of our algorithm. The obtained results demonstrate that our method is fairly robust and significantly outperforms other methods targeted at the same problem, in particular an expert system and an AdaBoost classifier.
Original languageEnglish
Title of host publicationProceedings of the Workshop of the Austrian Association for Pattern Recognition (OAGM/AAPR)
PublisherÖsterreichische Computergesellschaft
Publication statusPublished - 2009

Fields of Expertise

  • Information, Communication & Computing


Dive into the research topics of 'An Automatic Hybrid Segmentation Approach for Aligned Face Portrait Images'. Together they form a unique fingerprint.

Cite this