Research

Background Subtraction with Dirichlet Process Mixture Models


Reference:

Fincham Haines, T. and Xiang, T., 2013. Background Subtraction with Dirichlet Process Mixture Models. IEEE Transactions on Pattern Analysis and Machine Intelligence, 36 (4), pp. 670-683.

Related documents:

[img]
Preview
PDF (dp_bgs_j) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (4MB) | Preview

    Official URL:

    https://doi.org/10.1109/TPAMI.2013.239

    Abstract

    Video analysis often begins with background subtraction. This problem is often approached in two steps - a background model followed by a regularisation scheme. A model of the background allows it to be distinguished on a per-pixel basis from the foreground, whilst the regularisation combines information from adjacent pixels. We present a new method based on Dirichlet process Gaussian mixture models, which are used to estimate per-pixel background distributions. It is followed by probabilistic regularisation. Using a non-parametric Bayesian method allows per-pixel mode counts to be automatically inferred, avoiding over-/under- fitting. We also develop novel model learning algorithms for continuous update of the model in a principled fashion as the scene changes. These key advantages enable us to outperform the state-of-the-art alternatives on four benchmarks.

    Details

    Item Type Articles
    CreatorsFincham Haines, T.and Xiang, T.
    DOI10.1109/TPAMI.2013.239
    DepartmentsFaculty of Science > Computer Science
    Research CentresEPSRC Centre for Doctoral Training in Statistical Mathematics (SAMBa)
    Publisher Statementdp_bgs_j.pdf: © 2013 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/ republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works.
    RefereedYes
    StatusPublished
    ID Code56736

    Export

    Actions (login required)

    View Item

    Document Downloads

    More statistics for this item...