Video scene categorization by 3D hierarchical histogram matching


Gupta, P., Arrabolu, S. S., Brown, M. and Savarese, S., 2009. Video scene categorization by 3D hierarchical histogram matching. In: ICCV 2009: IEEE 12th International Conference on Computer Vision, 2009-09-29 - 2009-10-02.

Related documents:

PDF (Brown_iccv_2009.pdf) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (7MB) | Preview

    Official URL:

    Related URLs:


    In this paper we present a new method for categorizing video sequences capturing different scene classes. This can be seen as a generalization of previous work on scene classification from single images. A scene is represented by a collection of 3D points with an appearance based codeword attached to each point. The cloud of points is recovered by using a robust SFM algorithm applied on the video sequence. A hierarchical structure of histograms located at different locations and at different scales is used to capture the typical spatial distribution of 3D points and codewords in the working volume. The scene is classified by SVM equipped with a histogram matching kernel, similar to [21, 10, 16]. Results on a challenging dataset of 5 scene categories show competitive classification accuracy and superior performance with respect to a state-of-the-art 2D pyramid matching methods [16] applied to individual image frames.


    Item Type Conference or Workshop Items (Paper)
    CreatorsGupta, P., Arrabolu, S. S., Brown, M. and Savarese, S.
    Related URLs
    URLURL Type
    DepartmentsFaculty of Science > Computer Science
    Publisher StatementBrown_iccv_2009.pdf: © 2009 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, 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 component of this work in other works
    ID Code26113


    Actions (login required)

    View Item

    Document Downloads

    More statistics for this item...