Improving the Accuracy of Intelligent Pose Estimation Systems Through Low Level Image Processing Operations

Jannik Christian Lærkegård Pedersen, Mattias Foltmar Sander, Niklas Fruerlund Jensen, Jonas Lasham Lakhrissi, Mikkel Gede Hansen, Patrick Staalbo, Andreas Wulff-Abramsson

Research output: Contribution to journalConference article in JournalResearchpeer-review

Abstract

The development of powerful and popular machine learning driven pose estimation systems have been on the rise during the past years. In this research we have investigated how the accuracy level can be increased by applying low level image processing techniques unto the footage before they are submitted to the pose estimation system. The techniques used were high and low contrast, histogram equalization, sharpness and canny edge detection. By applying them on datasets, containing different environments and lighting conditions the system’s accuracy was increased, ranging from 0.29% increase to 38.37% increase dependent on the context. These increases have potential to upgrade the pose estimation system to be less lighting sensitive.
Original languageDanish
Journalforeløbig ukendt
Pages (from-to)1
Number of pages4
Publication statusAccepted/In press - 17 Jan 2019
EventDigital Image and Signal Processing - St. Hugh's College, Oxford University, Oxford, United Kingdom
Duration: 29 Apr 201930 Apr 2019
https://www.disp-conference.org/

Conference

ConferenceDigital Image and Signal Processing
LocationSt. Hugh's College, Oxford University
CountryUnited Kingdom
CityOxford
Period29/04/201930/04/2019
Internet address

Keywords

    Cite this

    Pedersen, J. C. L., Sander, M. F., Jensen, N. F., Lakhrissi, J. L., Hansen, M. G., Staalbo, P., & Wulff-Abramsson, A. (Accepted/In press). Improving the Accuracy of Intelligent Pose Estimation Systems Through Low Level Image Processing Operations. foreløbig ukendt, 1.
    Pedersen, Jannik Christian Lærkegård ; Sander, Mattias Foltmar ; Jensen, Niklas Fruerlund ; Lakhrissi, Jonas Lasham ; Hansen, Mikkel Gede ; Staalbo, Patrick ; Wulff-Abramsson, Andreas. / Improving the Accuracy of Intelligent Pose Estimation Systems Through Low Level Image Processing Operations. In: foreløbig ukendt. 2019 ; pp. 1.
    @inproceedings{a8cae45815494e26a7347be43d959cb6,
    title = "Improving the Accuracy of Intelligent Pose Estimation Systems Through Low Level Image Processing Operations",
    abstract = "The development of powerful and popular machine learning driven pose estimation systems have been on the rise during the past years. In this research we have investigated how the accuracy level can be increased by applying low level image processing techniques unto the footage before they are submitted to the pose estimation system. The techniques used were high and low contrast, histogram equalization, sharpness and canny edge detection. By applying them on datasets, containing different environments and lighting conditions the system’s accuracy was increased, ranging from 0.29{\%} increase to 38.37{\%} increase dependent on the context. These increases have potential to upgrade the pose estimation system to be less lighting sensitive.",
    keywords = "OpenPose, Image processing, Limb Estimation, Histogram equalization, Low-.level operations, Image Contrast, Sharpness, Canny edge detection",
    author = "Pedersen, {Jannik Christian L{\ae}rkeg{\aa}rd} and Sander, {Mattias Foltmar} and Jensen, {Niklas Fruerlund} and Lakhrissi, {Jonas Lasham} and Hansen, {Mikkel Gede} and Patrick Staalbo and Andreas Wulff-Abramsson",
    year = "2019",
    month = "1",
    day = "17",
    language = "Dansk",
    pages = "1",
    journal = "forel{\o}big ukendt",

    }

    Improving the Accuracy of Intelligent Pose Estimation Systems Through Low Level Image Processing Operations. / Pedersen, Jannik Christian Lærkegård ; Sander, Mattias Foltmar; Jensen, Niklas Fruerlund; Lakhrissi, Jonas Lasham ; Hansen, Mikkel Gede; Staalbo, Patrick ; Wulff-Abramsson, Andreas.

    In: foreløbig ukendt, 17.01.2019, p. 1.

    Research output: Contribution to journalConference article in JournalResearchpeer-review

    TY - GEN

    T1 - Improving the Accuracy of Intelligent Pose Estimation Systems Through Low Level Image Processing Operations

    AU - Pedersen, Jannik Christian Lærkegård

    AU - Sander, Mattias Foltmar

    AU - Jensen, Niklas Fruerlund

    AU - Lakhrissi, Jonas Lasham

    AU - Hansen, Mikkel Gede

    AU - Staalbo, Patrick

    AU - Wulff-Abramsson, Andreas

    PY - 2019/1/17

    Y1 - 2019/1/17

    N2 - The development of powerful and popular machine learning driven pose estimation systems have been on the rise during the past years. In this research we have investigated how the accuracy level can be increased by applying low level image processing techniques unto the footage before they are submitted to the pose estimation system. The techniques used were high and low contrast, histogram equalization, sharpness and canny edge detection. By applying them on datasets, containing different environments and lighting conditions the system’s accuracy was increased, ranging from 0.29% increase to 38.37% increase dependent on the context. These increases have potential to upgrade the pose estimation system to be less lighting sensitive.

    AB - The development of powerful and popular machine learning driven pose estimation systems have been on the rise during the past years. In this research we have investigated how the accuracy level can be increased by applying low level image processing techniques unto the footage before they are submitted to the pose estimation system. The techniques used were high and low contrast, histogram equalization, sharpness and canny edge detection. By applying them on datasets, containing different environments and lighting conditions the system’s accuracy was increased, ranging from 0.29% increase to 38.37% increase dependent on the context. These increases have potential to upgrade the pose estimation system to be less lighting sensitive.

    KW - OpenPose

    KW - Image processing

    KW - Limb Estimation

    KW - Histogram equalization

    KW - Low-.level operations

    KW - Image Contrast

    KW - Sharpness

    KW - Canny edge detection

    M3 - Konferenceartikel i tidsskrift

    SP - 1

    JO - foreløbig ukendt

    JF - foreløbig ukendt

    ER -

    Pedersen JCL, Sander MF, Jensen NF, Lakhrissi JL, Hansen MG, Staalbo P et al. Improving the Accuracy of Intelligent Pose Estimation Systems Through Low Level Image Processing Operations. foreløbig ukendt. 2019 Jan 17;1.