Multi-band Modelling of Appearance

Mikkel Bille Stegmann, Rasmus Larsen

    Research output: Contribution to journalConference articleResearchpeer-review

    Abstract

    Earlier work has demonstrated generative models capable of synthesising near photo-realistic grey-scale images of objects. These models have been augmented with colour information, and recently with edge information. This paper extends the active appearance model framework by modelling the appearance of both derived feature bands and an intensity band. As a special case of feature-band augmented appearance modelling we propose a dedicated representation with applications to face segmentation. The representation addresses a major problem within face recognition by lowering the sensitivity to lighting conditions. Results show that the localisation accuracy of facial features is considerably increased using this appearance representation under diffuse and directional lighting and at multiple scales.
    Original languageEnglish
    JournalImage and Vision Computing
    Volume21
    Issue number1
    Pages (from-to)61-67
    ISSN0262-8856
    DOIs
    Publication statusPublished - 2003
    Event1st International Workshop on Generative-Model-Based Vision - Copenhagen, Denmark
    Duration: 2 Jun 2002 → …
    Conference number: 1

    Conference

    Conference1st International Workshop on Generative-Model-Based Vision
    Number1
    CountryDenmark
    CityCopenhagen
    Period02/06/2002 → …

    Keywords

    • Generative model
    • Active appearance model
    • Lighting invariance
    • Face recognition
    • Segmentation

    Fingerprint Dive into the research topics of 'Multi-band Modelling of Appearance'. Together they form a unique fingerprint.

    Cite this