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 conference without publisher/journalPaper without publisher/journalResearch


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 languageEnglish
Publication date4 May 2019
Number of pages4
Publication statusPublished - 4 May 2019
EventInternational Conference on Digital Image & Signal Processing (DISP’19) - Oxford, United Kingdom
Duration: 29 Apr 201930 Apr 2019


ConferenceInternational Conference on Digital Image & Signal Processing (DISP’19)
Country/TerritoryUnited Kingdom

Cite this