Learning algorithms and frame signatures for video similarity ranking

Teruki Horie, Akihiro Shikano, Hiromichi Iwase, Yasuo Matsuyama

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    1 Citation (Scopus)

    Abstract

    Learning algorithms that harmonize standardized video similarity tools and an integrated system are presented. The learning algorithms extract exemplars reflecting time courses of video frames. There were five types of such clustering methods. Among them, this paper chooses a method called time-partition pairwise nearest-neighbor because of its reduced complexity. On the similarity comparison among videos whose lengths vary, the M-distance that can absorb the difference of the exemplar cardinalities is utilized both for global and local matching. Given the order-aware clustering and the M-distance comparison, system designers can build a basic similar-video retrieval system. This paper promotes further enhancement on the exemplar similarity that matches the video signature tools for the multimedia content description interface by ISO/IEC. This development showed the ability of the similarity ranking together with the detection of plagiarism of video scenes. Precision-recall curves showed a high performance in this experiment.

    Original languageEnglish
    Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    PublisherSpringer Verlag
    Pages147-157
    Number of pages11
    Volume9489
    ISBN (Print)9783319265315
    DOIs
    Publication statusPublished - 2015
    Event22nd International Conference on Neural Information Processing, ICONIP 2015 - Istanbul, Turkey
    Duration: 2015 Nov 92015 Nov 12

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume9489
    ISSN (Print)03029743
    ISSN (Electronic)16113349

    Other

    Other22nd International Conference on Neural Information Processing, ICONIP 2015
    CountryTurkey
    CityIstanbul
    Period15/11/915/11/12

    Fingerprint

    Learning algorithms
    Learning Algorithm
    Ranking
    Signature
    Video Retrieval
    Integrated System
    Clustering Methods
    Multimedia
    Pairwise
    Cardinality
    Nearest Neighbor
    High Performance
    Enhancement
    Choose
    Experiments
    Partition
    Clustering
    Vary
    Curve
    Similarity

    Keywords

    • Exemplar
    • Frame signature
    • M-distance
    • Numerical label
    • Video similarity ranking

    ASJC Scopus subject areas

    • Computer Science(all)
    • Theoretical Computer Science

    Cite this

    Horie, T., Shikano, A., Iwase, H., & Matsuyama, Y. (2015). Learning algorithms and frame signatures for video similarity ranking. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9489, pp. 147-157). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9489). Springer Verlag. https://doi.org/10.1007/978-3-319-26532-2_17

    Learning algorithms and frame signatures for video similarity ranking. / Horie, Teruki; Shikano, Akihiro; Iwase, Hiromichi; Matsuyama, Yasuo.

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 9489 Springer Verlag, 2015. p. 147-157 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9489).

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Horie, T, Shikano, A, Iwase, H & Matsuyama, Y 2015, Learning algorithms and frame signatures for video similarity ranking. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 9489, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9489, Springer Verlag, pp. 147-157, 22nd International Conference on Neural Information Processing, ICONIP 2015, Istanbul, Turkey, 15/11/9. https://doi.org/10.1007/978-3-319-26532-2_17
    Horie T, Shikano A, Iwase H, Matsuyama Y. Learning algorithms and frame signatures for video similarity ranking. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 9489. Springer Verlag. 2015. p. 147-157. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-26532-2_17
    Horie, Teruki ; Shikano, Akihiro ; Iwase, Hiromichi ; Matsuyama, Yasuo. / Learning algorithms and frame signatures for video similarity ranking. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 9489 Springer Verlag, 2015. pp. 147-157 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
    @inproceedings{2fda4989fde84d62ae311d38a1bf2bfb,
    title = "Learning algorithms and frame signatures for video similarity ranking",
    abstract = "Learning algorithms that harmonize standardized video similarity tools and an integrated system are presented. The learning algorithms extract exemplars reflecting time courses of video frames. There were five types of such clustering methods. Among them, this paper chooses a method called time-partition pairwise nearest-neighbor because of its reduced complexity. On the similarity comparison among videos whose lengths vary, the M-distance that can absorb the difference of the exemplar cardinalities is utilized both for global and local matching. Given the order-aware clustering and the M-distance comparison, system designers can build a basic similar-video retrieval system. This paper promotes further enhancement on the exemplar similarity that matches the video signature tools for the multimedia content description interface by ISO/IEC. This development showed the ability of the similarity ranking together with the detection of plagiarism of video scenes. Precision-recall curves showed a high performance in this experiment.",
    keywords = "Exemplar, Frame signature, M-distance, Numerical label, Video similarity ranking",
    author = "Teruki Horie and Akihiro Shikano and Hiromichi Iwase and Yasuo Matsuyama",
    year = "2015",
    doi = "10.1007/978-3-319-26532-2_17",
    language = "English",
    isbn = "9783319265315",
    volume = "9489",
    series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
    publisher = "Springer Verlag",
    pages = "147--157",
    booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

    }

    TY - GEN

    T1 - Learning algorithms and frame signatures for video similarity ranking

    AU - Horie, Teruki

    AU - Shikano, Akihiro

    AU - Iwase, Hiromichi

    AU - Matsuyama, Yasuo

    PY - 2015

    Y1 - 2015

    N2 - Learning algorithms that harmonize standardized video similarity tools and an integrated system are presented. The learning algorithms extract exemplars reflecting time courses of video frames. There were five types of such clustering methods. Among them, this paper chooses a method called time-partition pairwise nearest-neighbor because of its reduced complexity. On the similarity comparison among videos whose lengths vary, the M-distance that can absorb the difference of the exemplar cardinalities is utilized both for global and local matching. Given the order-aware clustering and the M-distance comparison, system designers can build a basic similar-video retrieval system. This paper promotes further enhancement on the exemplar similarity that matches the video signature tools for the multimedia content description interface by ISO/IEC. This development showed the ability of the similarity ranking together with the detection of plagiarism of video scenes. Precision-recall curves showed a high performance in this experiment.

    AB - Learning algorithms that harmonize standardized video similarity tools and an integrated system are presented. The learning algorithms extract exemplars reflecting time courses of video frames. There were five types of such clustering methods. Among them, this paper chooses a method called time-partition pairwise nearest-neighbor because of its reduced complexity. On the similarity comparison among videos whose lengths vary, the M-distance that can absorb the difference of the exemplar cardinalities is utilized both for global and local matching. Given the order-aware clustering and the M-distance comparison, system designers can build a basic similar-video retrieval system. This paper promotes further enhancement on the exemplar similarity that matches the video signature tools for the multimedia content description interface by ISO/IEC. This development showed the ability of the similarity ranking together with the detection of plagiarism of video scenes. Precision-recall curves showed a high performance in this experiment.

    KW - Exemplar

    KW - Frame signature

    KW - M-distance

    KW - Numerical label

    KW - Video similarity ranking

    UR - http://www.scopus.com/inward/record.url?scp=84952802973&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84952802973&partnerID=8YFLogxK

    U2 - 10.1007/978-3-319-26532-2_17

    DO - 10.1007/978-3-319-26532-2_17

    M3 - Conference contribution

    AN - SCOPUS:84952802973

    SN - 9783319265315

    VL - 9489

    T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

    SP - 147

    EP - 157

    BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

    PB - Springer Verlag

    ER -