Hierarchical sparse spectral clustering for image set classification

Arif Mahmood, Ajmal Mian

    Research output: Chapter in Book/Conference paperConference paper

    5 Citations (Scopus)

    Abstract

    We present a structural matching technique for robust classification based on image sets. In set based classification, a probe set is matched with a number of gallery sets and assigned the label of the most similar set. We represent each image set by a sparse dictionary and compute a similarity matrix by matching all the dictionary atoms of the gallery and probe sets. The similarity matrix comprises the sparse coding coefficients and forms a fully connected directed graph. The nodes of the graph are the dictionary atoms and the edges are the sparse coefficients. The graph is converted to an undirected graph with positive edge weights and spectral clustering is used to cut the graph into two balanced partitions using the normalized cut algorithm. This process is repeated until the graph reduces to critical and non-critical partitions. A critical partition contains atoms with the same gallery label along with one or more probe atoms whereas a non-critical partition either consists of only probe atoms or atoms with multiple gallery labels with no probe atom. Using the critical partitions, we define a novel set based similarity measure and assign the probe set the label of the gallery set with maximum similarity. The proposed algorithm is applied to image set based face recognition using two standard databases. Comparison with existing techniques shows the validity and robustness of our algorithm in the presence of outlier images.
    Original languageEnglish
    Title of host publicationProceedings of the British Machine Vision Conference 2012
    Place of PublicationUK
    PublisherBMVA Press
    Pages1-11
    Volume1
    ISBN (Print)1901725464
    DOIs
    Publication statusPublished - 2012
    EventHierarchical sparse spectral clustering for image set classification - UK
    Duration: 1 Jan 2012 → …

    Conference

    ConferenceHierarchical sparse spectral clustering for image set classification
    Period1/01/12 → …

    Fingerprint

    Atoms
    Labels
    Glossaries
    Directed graphs
    Face recognition

    Cite this

    Mahmood, A., & Mian, A. (2012). Hierarchical sparse spectral clustering for image set classification. In Proceedings of the British Machine Vision Conference 2012 (Vol. 1, pp. 1-11). UK: BMVA Press. https://doi.org/10.5244/C.26.51
    Mahmood, Arif ; Mian, Ajmal. / Hierarchical sparse spectral clustering for image set classification. Proceedings of the British Machine Vision Conference 2012. Vol. 1 UK : BMVA Press, 2012. pp. 1-11
    @inproceedings{add66f4b769f49d3860ad928b81917b6,
    title = "Hierarchical sparse spectral clustering for image set classification",
    abstract = "We present a structural matching technique for robust classification based on image sets. In set based classification, a probe set is matched with a number of gallery sets and assigned the label of the most similar set. We represent each image set by a sparse dictionary and compute a similarity matrix by matching all the dictionary atoms of the gallery and probe sets. The similarity matrix comprises the sparse coding coefficients and forms a fully connected directed graph. The nodes of the graph are the dictionary atoms and the edges are the sparse coefficients. The graph is converted to an undirected graph with positive edge weights and spectral clustering is used to cut the graph into two balanced partitions using the normalized cut algorithm. This process is repeated until the graph reduces to critical and non-critical partitions. A critical partition contains atoms with the same gallery label along with one or more probe atoms whereas a non-critical partition either consists of only probe atoms or atoms with multiple gallery labels with no probe atom. Using the critical partitions, we define a novel set based similarity measure and assign the probe set the label of the gallery set with maximum similarity. The proposed algorithm is applied to image set based face recognition using two standard databases. Comparison with existing techniques shows the validity and robustness of our algorithm in the presence of outlier images.",
    author = "Arif Mahmood and Ajmal Mian",
    year = "2012",
    doi = "10.5244/C.26.51",
    language = "English",
    isbn = "1901725464",
    volume = "1",
    pages = "1--11",
    booktitle = "Proceedings of the British Machine Vision Conference 2012",
    publisher = "BMVA Press",

    }

    Mahmood, A & Mian, A 2012, Hierarchical sparse spectral clustering for image set classification. in Proceedings of the British Machine Vision Conference 2012. vol. 1, BMVA Press, UK, pp. 1-11, Hierarchical sparse spectral clustering for image set classification, 1/01/12. https://doi.org/10.5244/C.26.51

    Hierarchical sparse spectral clustering for image set classification. / Mahmood, Arif; Mian, Ajmal.

    Proceedings of the British Machine Vision Conference 2012. Vol. 1 UK : BMVA Press, 2012. p. 1-11.

    Research output: Chapter in Book/Conference paperConference paper

    TY - GEN

    T1 - Hierarchical sparse spectral clustering for image set classification

    AU - Mahmood, Arif

    AU - Mian, Ajmal

    PY - 2012

    Y1 - 2012

    N2 - We present a structural matching technique for robust classification based on image sets. In set based classification, a probe set is matched with a number of gallery sets and assigned the label of the most similar set. We represent each image set by a sparse dictionary and compute a similarity matrix by matching all the dictionary atoms of the gallery and probe sets. The similarity matrix comprises the sparse coding coefficients and forms a fully connected directed graph. The nodes of the graph are the dictionary atoms and the edges are the sparse coefficients. The graph is converted to an undirected graph with positive edge weights and spectral clustering is used to cut the graph into two balanced partitions using the normalized cut algorithm. This process is repeated until the graph reduces to critical and non-critical partitions. A critical partition contains atoms with the same gallery label along with one or more probe atoms whereas a non-critical partition either consists of only probe atoms or atoms with multiple gallery labels with no probe atom. Using the critical partitions, we define a novel set based similarity measure and assign the probe set the label of the gallery set with maximum similarity. The proposed algorithm is applied to image set based face recognition using two standard databases. Comparison with existing techniques shows the validity and robustness of our algorithm in the presence of outlier images.

    AB - We present a structural matching technique for robust classification based on image sets. In set based classification, a probe set is matched with a number of gallery sets and assigned the label of the most similar set. We represent each image set by a sparse dictionary and compute a similarity matrix by matching all the dictionary atoms of the gallery and probe sets. The similarity matrix comprises the sparse coding coefficients and forms a fully connected directed graph. The nodes of the graph are the dictionary atoms and the edges are the sparse coefficients. The graph is converted to an undirected graph with positive edge weights and spectral clustering is used to cut the graph into two balanced partitions using the normalized cut algorithm. This process is repeated until the graph reduces to critical and non-critical partitions. A critical partition contains atoms with the same gallery label along with one or more probe atoms whereas a non-critical partition either consists of only probe atoms or atoms with multiple gallery labels with no probe atom. Using the critical partitions, we define a novel set based similarity measure and assign the probe set the label of the gallery set with maximum similarity. The proposed algorithm is applied to image set based face recognition using two standard databases. Comparison with existing techniques shows the validity and robustness of our algorithm in the presence of outlier images.

    U2 - 10.5244/C.26.51

    DO - 10.5244/C.26.51

    M3 - Conference paper

    SN - 1901725464

    VL - 1

    SP - 1

    EP - 11

    BT - Proceedings of the British Machine Vision Conference 2012

    PB - BMVA Press

    CY - UK

    ER -

    Mahmood A, Mian A. Hierarchical sparse spectral clustering for image set classification. In Proceedings of the British Machine Vision Conference 2012. Vol. 1. UK: BMVA Press. 2012. p. 1-11 https://doi.org/10.5244/C.26.51