Human tracking method based on improved HOG+Real AdaBoost

Daisuke Aoki, Junzo Watada

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

    Abstract

    This paper proposes an object detection method that uses Histograms of Oriented Gradients (HOG) features using boosting algorithm. There has been done many research works in late years on statistical learning methods and object detection methods that associate low level of features obtained. However the proposed approach, low level of HOG features are associated by using Real AdaBoost to continuously achieve features. In this wise, it is possible to capture a shape of edge continuity, which single HOG features can't do, so highly accuracy detection is realized. This paper, to evaluate the effectiveness of the proposed method, three different experiments with different patterns are conducted for detecting humans. Moreover, a boosting classifier is used to represent the co-occurrence of HOG features appearance for detecting a human.

    Original languageEnglish
    Title of host publication2015 10th Asian Control Conference: Emerging Control Techniques for a Sustainable World, ASCC 2015
    PublisherInstitute of Electrical and Electronics Engineers Inc.
    ISBN (Print)9781479978625
    DOIs
    Publication statusPublished - 2015 Sep 8
    Event10th Asian Control Conference, ASCC 2015 - Kota Kinabalu, Malaysia
    Duration: 2015 May 312015 Jun 3

    Other

    Other10th Asian Control Conference, ASCC 2015
    CountryMalaysia
    CityKota Kinabalu
    Period15/5/3115/6/3

    Fingerprint

    Adaptive boosting
    Classifiers
    Experiments
    Object detection

    Keywords

    • Histograms of Oriented Gradients
    • Real Adaboost

    ASJC Scopus subject areas

    • Control and Systems Engineering

    Cite this

    Aoki, D., & Watada, J. (2015). Human tracking method based on improved HOG+Real AdaBoost. In 2015 10th Asian Control Conference: Emerging Control Techniques for a Sustainable World, ASCC 2015 [7244780] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ASCC.2015.7244780

    Human tracking method based on improved HOG+Real AdaBoost. / Aoki, Daisuke; Watada, Junzo.

    2015 10th Asian Control Conference: Emerging Control Techniques for a Sustainable World, ASCC 2015. Institute of Electrical and Electronics Engineers Inc., 2015. 7244780.

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

    Aoki, D & Watada, J 2015, Human tracking method based on improved HOG+Real AdaBoost. in 2015 10th Asian Control Conference: Emerging Control Techniques for a Sustainable World, ASCC 2015., 7244780, Institute of Electrical and Electronics Engineers Inc., 10th Asian Control Conference, ASCC 2015, Kota Kinabalu, Malaysia, 15/5/31. https://doi.org/10.1109/ASCC.2015.7244780
    Aoki D, Watada J. Human tracking method based on improved HOG+Real AdaBoost. In 2015 10th Asian Control Conference: Emerging Control Techniques for a Sustainable World, ASCC 2015. Institute of Electrical and Electronics Engineers Inc. 2015. 7244780 https://doi.org/10.1109/ASCC.2015.7244780
    Aoki, Daisuke ; Watada, Junzo. / Human tracking method based on improved HOG+Real AdaBoost. 2015 10th Asian Control Conference: Emerging Control Techniques for a Sustainable World, ASCC 2015. Institute of Electrical and Electronics Engineers Inc., 2015.
    @inproceedings{510e62dc7b434596a75f5621dc139f90,
    title = "Human tracking method based on improved HOG+Real AdaBoost",
    abstract = "This paper proposes an object detection method that uses Histograms of Oriented Gradients (HOG) features using boosting algorithm. There has been done many research works in late years on statistical learning methods and object detection methods that associate low level of features obtained. However the proposed approach, low level of HOG features are associated by using Real AdaBoost to continuously achieve features. In this wise, it is possible to capture a shape of edge continuity, which single HOG features can't do, so highly accuracy detection is realized. This paper, to evaluate the effectiveness of the proposed method, three different experiments with different patterns are conducted for detecting humans. Moreover, a boosting classifier is used to represent the co-occurrence of HOG features appearance for detecting a human.",
    keywords = "Histograms of Oriented Gradients, Real Adaboost",
    author = "Daisuke Aoki and Junzo Watada",
    year = "2015",
    month = "9",
    day = "8",
    doi = "10.1109/ASCC.2015.7244780",
    language = "English",
    isbn = "9781479978625",
    booktitle = "2015 10th Asian Control Conference: Emerging Control Techniques for a Sustainable World, ASCC 2015",
    publisher = "Institute of Electrical and Electronics Engineers Inc.",

    }

    TY - GEN

    T1 - Human tracking method based on improved HOG+Real AdaBoost

    AU - Aoki, Daisuke

    AU - Watada, Junzo

    PY - 2015/9/8

    Y1 - 2015/9/8

    N2 - This paper proposes an object detection method that uses Histograms of Oriented Gradients (HOG) features using boosting algorithm. There has been done many research works in late years on statistical learning methods and object detection methods that associate low level of features obtained. However the proposed approach, low level of HOG features are associated by using Real AdaBoost to continuously achieve features. In this wise, it is possible to capture a shape of edge continuity, which single HOG features can't do, so highly accuracy detection is realized. This paper, to evaluate the effectiveness of the proposed method, three different experiments with different patterns are conducted for detecting humans. Moreover, a boosting classifier is used to represent the co-occurrence of HOG features appearance for detecting a human.

    AB - This paper proposes an object detection method that uses Histograms of Oriented Gradients (HOG) features using boosting algorithm. There has been done many research works in late years on statistical learning methods and object detection methods that associate low level of features obtained. However the proposed approach, low level of HOG features are associated by using Real AdaBoost to continuously achieve features. In this wise, it is possible to capture a shape of edge continuity, which single HOG features can't do, so highly accuracy detection is realized. This paper, to evaluate the effectiveness of the proposed method, three different experiments with different patterns are conducted for detecting humans. Moreover, a boosting classifier is used to represent the co-occurrence of HOG features appearance for detecting a human.

    KW - Histograms of Oriented Gradients

    KW - Real Adaboost

    UR - http://www.scopus.com/inward/record.url?scp=84957655871&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84957655871&partnerID=8YFLogxK

    U2 - 10.1109/ASCC.2015.7244780

    DO - 10.1109/ASCC.2015.7244780

    M3 - Conference contribution

    SN - 9781479978625

    BT - 2015 10th Asian Control Conference: Emerging Control Techniques for a Sustainable World, ASCC 2015

    PB - Institute of Electrical and Electronics Engineers Inc.

    ER -