Skip to content

Research at St Andrews

An information-theoretic approach to face recognition from face motion manifolds

Research output: Contribution to journalArticlepeer-review

Author(s)

Oggie Arandelovic, Roberto Cipolla

School/Research organisations

Abstract

In this work, we consider face recognition from face motion manifolds (FMMs). The use of the resistor-average distance (RAD) as a dissimilarity measure between densities confined to FMMs is motivated in the proposed information-theoretic approach to modelling face appearance. We introduce a kernel-based algorithm that makes use of the simplicity of the closed-form expression for RAD between two Gaussian densities, while allowing for modelling of complex and nonlinear, but intrinsically low-dimensional manifolds. Additionally, it is shown how geodesically local FMM structure can be modelled, naturally leading to a stochastic algorithm for generalizing to unseen modes of data variation. Recognition performance of our method is demonstrated experimentally and is shown to exceed that of state-of-the-art algorithms. Recognition rate of 98% was achieved on a database of 100 people under varying illumination.

Close

Details

Original languageEnglish
Pages (from-to)639-647
Number of pages9
JournalImage and Vision Computing
Volume24
Issue number6
DOIs
Publication statusPublished - 1 Jun 2006

    Research areas

  • Face motion manifolds, Face recognition, Kernel, Resistor-average distance

Discover related content
Find related publications, people, projects and more using interactive charts.

View graph of relations

Related by author

  1. Review of automatic microexpression recognition in the past decade

    Zhang, L. & Arandjelović, O., 2 May 2021, In: Machine Learning and Knowledge Extraction. 3, 2, p. 414-434 21 p.

    Research output: Contribution to journalReview articlepeer-review

  2. Assessment of immunological features in muscle-invasive bladder cancer prognosis using ensemble learning

    Gavriel, C., Dimitriou, N., Brieu, N., Nearchou, I. P., Arandelovic, O., Schmidt, G., Harrison, D. J. & Caie, P. D., 1 Apr 2021, In: Cancers. 13, 7, 1624.

    Research output: Contribution to journalArticlepeer-review

  3. Corrigendum: Deep Learning for Whole Slide Image Analysis: An Overview (Front. Med. (2019), 6, (264), 10.3389/fmed.2019.00264)

    Dimitriou, N., Arandjelović, O. & Caie, P. D., 19 Aug 2020, In: Frontiers in Medicine. 7, 419.

    Research output: Contribution to journalComment/debatepeer-review

Related by journal

  1. Texture features for object salience

    Terzić, K., Krishna, S. & du Buf, J. M. H., Nov 2017, In: Image and Vision Computing. 67, p. 43-51

    Research output: Contribution to journalArticlepeer-review

  2. Digitization of non-regular shapes in arbitrary dimensions

    Stelldinger, P. & Terzic, K., 1 Oct 2008, In: Image and Vision Computing. 26, 10, p. 1338-1346 9 p.

    Research output: Contribution to journalArticlepeer-review

ID: 242483060

Top