At the heart of the so-called CEVA-MM3101 IP is a programmable Vector Processing (VP) engine. Based on a dedicated pixel-processing VLIW/SIMD architecture with a 10-stage pipeline, it contains seven different units that can work in parallel, and contains special instructions that can be configured to create proprietary filters for video and imaging processing.
The CEVA-MM3101 IP is supported by a software development toolkit that includes an optimizing C-compiler, IDE, debugger, simulators, and profiler. To aid vector processor optimization, the IDE provides a view into all core and memory resources.
As part of the solution, the IP includes a library of video and imaging functions, including image pipeline kernels, linear and nonlinear filters, pre- and post-processing functions, face detection, and video codec kernels.
To address the burgeoning area of embedded vision, CEVA has collaborated with Eyesight (Herzeliya, Israel) to develop a hand gesture recognition system based on the CEVA-MM3101 platform.
-- By Dave Wilson, Senior Editor, Vision Systems Design