A pipelined architecture for motion tracking on a multicore environment

Motion tracking is an important task in computer vision applications. Embedded system design for this task needs to address goals of high performance with low area and power requirements. In this paper, we propose a pipelined parallel architecture for motion tracking that is appropriate for implementation on a multicore environment. The architecture comprises of modules for frame subtraction, binarization, morphological operations and connected component analysis operating in sequence on an image frame. Successive frames are processed in a pipeline on multicores. Embedded system realization on a multicore XMOS microcontroller runs the drivers for interfacing image sensor, SDRAM and image sensor on different cores along with the various stages of the image processing pipeline. The realization achieves a frame rate of 8 fps for an image size of 480×272. Further, the solution is area-efficient and is based on a single XMOS sliceKIT with support (in the form of compact slices) for camera, LCD, memory and other units.