Master’s thesis: Parallel Computing for Digital Signal Processing on Mobile Device GPUs

2014-05-01 • gpgpu, imageproc, mobile development, android, opengl, opencl, renderscript

In my master’s thesis I analyzed the potentials of general purpose computing on graphics processing units (GPGPU) on mobile devices such as smartphones or tablets. The question was, if and how the GPU on such devices can be used to speed up certain algorithms especially in the fields of image processing. GPU computing technologies such as OpenCL, OpenGL shaders, and Android RenderScript are assessed in the thesis.

The abstract of my thesis reads as follows:

This thesis studies how certain popular algorithms in the field of image and audio processing can be accelerated on mobile devices by means of parallel execution on their graphics processing unit (GPU). Several technologies with which this can be achieved are compared in terms of possible performance improvements, hardware and software support, as well as limitations of their programming model and functionality. The results of this research are applied in a practical project, consisting of performance improvements for marker detection in an Augmented Reality application for mobile devices.

The PDF of my master’s thesis and the source-code for some prototypes are available for download: