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

Publikation: Bidrag til tidsskriftKonferenceartikel i tidsskriftForskningpeer review

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

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.
Luk

Detaljer

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.
OriginalsprogDansk
Tidsskriftforeløbig ukendt
Sider (fra-til)1
Antal sider4
StatusAccepteret/In press - 17 jan. 2019
PublikationsartForskning
Peer reviewJa
BegivenhedDigital Image and Signal Processing - St. Hugh's College, Oxford University, Oxford, Storbritannien
Varighed: 29 apr. 201930 apr. 2019
https://www.disp-conference.org/

Konference

KonferenceDigital Image and Signal Processing
LokationSt. Hugh's College, Oxford University
LandStorbritannien
ByOxford
Periode29/04/201930/04/2019
Internetadresse

    Forskningsområder

  • OpenPose, Image processing, Limb Estimation, Histogram equalization, Low-.level operations, Image Contrast, Sharpness, Canny edge detection
ID: 294188919