Mobile phone performance analysis for camera based visual interactions

摘要

Vision based technology such as motion detection has long been limited to the domain of powerful processor intensive systems such as desktop PC's and specialist hardware solutions. With the advent of much faster mobile phone processors and memory we are now seeing a plethora of feature rich software being deployed onto the mobile platform. Since these high powered smart phones are now equipped with cameras, it has become feasible to combine their powerful processors and the camera to support new ways of interacting with the phone. However, it is not clear whether or not these processor intensive visual interactions can in fact be run at an acceptable speed on current mobile handsets. In this paper we look at one of the most popular and widespread mobile smart phone systems; the Symbian s60 and benchmark the speed, accuracy and deployability of the three popular mobile languages. We test a pixel thresholding algorithm in, C++, Python and Java and rank them based on their speed within the contextof intensive image based processing.