Evaluation of SIFT on Android-based mobile devices and personal computers

  • Ian Cornelius

    Student thesis: Master's ThesisMaster of Science by Research


    The purpose of this paper is to evaluate two algorithms for the feasibility of object recognition on mobile devices. The majority of this paper is based upon the work by Lowe (1999, 2004) with the SIFT algorithm and a discussion regarding the DAISY algorithm by Tola et al. (2010), evaluating the key aspects of the algorithm and whether it could be a used on mobile devices.

    The SIFT algorithm has been implemented to run on a desktop machine and an Android-based mobile device to test the feasibility of object detection. The implementation for the desktop was done for a comparative study between the mobile device and the desktop computer. Source-code of the implementation on the desktop and the Android platform can be found at: https://github.com/iancornelius/SIFT-Android.

    The results from the outcome found that the task of object recognition on mobile devices is feasible. Further work to be carried out would be the implementation and evaluation of DAISY on the desktop and Android mobile-device to compare whether DAISY can improve the results gained on single-core devices. Another aspect could be how threading would affect the results especially with the computation on the extraction of descriptors from images being done simulataneously.
    Date of Award2014
    Original languageEnglish
    Awarding Institution
    • Coventry University


    • SIFT
    • Personal Computers
    • Mobile Devices
    • Android

    Cite this