A novel method for object tracking and segmentation using online Hough forests and convex relaxation

Zhengjian Kang, Edward K. Wong

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

    Abstract

    We propose a novel method for object tracking and segmentation by using online Hough forests and convex relaxation. Our method extracts object contour during tracking rather than using a bounding box or an ellipse to locate the object. Unlike conventional active contour methods that use consistent intensity or color distribution as constraints, our method uses Hough forests for online discriminative learning, resulting in faster convergence and more accurate segmentation. We use Bayesian formulation to model the probability of the contour, given the description of the regions and the edges. Additionally, the Hough forests provide an estimate of the initial location of the object to improve accuracy. Segmentation is then formulated as a convex relaxation optimization problem. Experimental results show the effectiveness and robustness of our method. The results also show that our method outperforms some of the state-of-the-art methods.

    Original languageEnglish (US)
    Title of host publication2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings
    Pages3870-3874
    Number of pages5
    DOIs
    StatePublished - 2013
    Event2013 20th IEEE International Conference on Image Processing, ICIP 2013 - Melbourne, VIC, Australia
    Duration: Sep 15 2013Sep 18 2013

    Other

    Other2013 20th IEEE International Conference on Image Processing, ICIP 2013
    CountryAustralia
    CityMelbourne, VIC
    Period9/15/139/18/13

    Fingerprint

    Color

    Keywords

    • convex relaxation
    • Hough forests
    • Object tracking
    • segmentation

    ASJC Scopus subject areas

    • Computer Vision and Pattern Recognition

    Cite this

    Kang, Z., & Wong, E. K. (2013). A novel method for object tracking and segmentation using online Hough forests and convex relaxation. In 2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings (pp. 3870-3874). [6738797] https://doi.org/10.1109/ICIP.2013.6738797

    A novel method for object tracking and segmentation using online Hough forests and convex relaxation. / Kang, Zhengjian; Wong, Edward K.

    2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings. 2013. p. 3870-3874 6738797.

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

    Kang, Z & Wong, EK 2013, A novel method for object tracking and segmentation using online Hough forests and convex relaxation. in 2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings., 6738797, pp. 3870-3874, 2013 20th IEEE International Conference on Image Processing, ICIP 2013, Melbourne, VIC, Australia, 9/15/13. https://doi.org/10.1109/ICIP.2013.6738797
    Kang Z, Wong EK. A novel method for object tracking and segmentation using online Hough forests and convex relaxation. In 2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings. 2013. p. 3870-3874. 6738797 https://doi.org/10.1109/ICIP.2013.6738797
    Kang, Zhengjian ; Wong, Edward K. / A novel method for object tracking and segmentation using online Hough forests and convex relaxation. 2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings. 2013. pp. 3870-3874
    @inproceedings{241f96a4807e4cf6b095b8df06f401b4,
    title = "A novel method for object tracking and segmentation using online Hough forests and convex relaxation",
    abstract = "We propose a novel method for object tracking and segmentation by using online Hough forests and convex relaxation. Our method extracts object contour during tracking rather than using a bounding box or an ellipse to locate the object. Unlike conventional active contour methods that use consistent intensity or color distribution as constraints, our method uses Hough forests for online discriminative learning, resulting in faster convergence and more accurate segmentation. We use Bayesian formulation to model the probability of the contour, given the description of the regions and the edges. Additionally, the Hough forests provide an estimate of the initial location of the object to improve accuracy. Segmentation is then formulated as a convex relaxation optimization problem. Experimental results show the effectiveness and robustness of our method. The results also show that our method outperforms some of the state-of-the-art methods.",
    keywords = "convex relaxation, Hough forests, Object tracking, segmentation",
    author = "Zhengjian Kang and Wong, {Edward K.}",
    year = "2013",
    doi = "10.1109/ICIP.2013.6738797",
    language = "English (US)",
    isbn = "9781479923410",
    pages = "3870--3874",
    booktitle = "2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings",

    }

    TY - GEN

    T1 - A novel method for object tracking and segmentation using online Hough forests and convex relaxation

    AU - Kang, Zhengjian

    AU - Wong, Edward K.

    PY - 2013

    Y1 - 2013

    N2 - We propose a novel method for object tracking and segmentation by using online Hough forests and convex relaxation. Our method extracts object contour during tracking rather than using a bounding box or an ellipse to locate the object. Unlike conventional active contour methods that use consistent intensity or color distribution as constraints, our method uses Hough forests for online discriminative learning, resulting in faster convergence and more accurate segmentation. We use Bayesian formulation to model the probability of the contour, given the description of the regions and the edges. Additionally, the Hough forests provide an estimate of the initial location of the object to improve accuracy. Segmentation is then formulated as a convex relaxation optimization problem. Experimental results show the effectiveness and robustness of our method. The results also show that our method outperforms some of the state-of-the-art methods.

    AB - We propose a novel method for object tracking and segmentation by using online Hough forests and convex relaxation. Our method extracts object contour during tracking rather than using a bounding box or an ellipse to locate the object. Unlike conventional active contour methods that use consistent intensity or color distribution as constraints, our method uses Hough forests for online discriminative learning, resulting in faster convergence and more accurate segmentation. We use Bayesian formulation to model the probability of the contour, given the description of the regions and the edges. Additionally, the Hough forests provide an estimate of the initial location of the object to improve accuracy. Segmentation is then formulated as a convex relaxation optimization problem. Experimental results show the effectiveness and robustness of our method. The results also show that our method outperforms some of the state-of-the-art methods.

    KW - convex relaxation

    KW - Hough forests

    KW - Object tracking

    KW - segmentation

    UR - http://www.scopus.com/inward/record.url?scp=84897798876&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84897798876&partnerID=8YFLogxK

    U2 - 10.1109/ICIP.2013.6738797

    DO - 10.1109/ICIP.2013.6738797

    M3 - Conference contribution

    SN - 9781479923410

    SP - 3870

    EP - 3874

    BT - 2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings

    ER -