The EV52 features a dual-core 32-bit CPU with a programmable CNN object detection engine that is user configurable with up to eight processing elements. The EV54 features a quad-core CPU and the programmable CNN object detection engine.
To speed application software development, the EV5x Processor family is supported by a comprehensive software programming environment based on existing and emerging embedded vision standards including OpenCV and OpenVX™, as well as Synopsys’ ARC MetaWare Development Toolkit.
The OpenCV source libraries available for EV Processors provide more than 2500 functions for real-time computer vision. The processors are programmable and can be trained to support any object detection graph. The OpenVX framework includes 43 standard computer vision kernels that have been optimized to run on the EV Processors, such as edge detection, image pyramid creation and optical flow estimation. Users can also define new OpenVX kernels, giving them flexibility for their current vision applications and the ability to address future object detection requirements. The OpenVX runtime distributes tiled kernel execution over the EV Processors’ multiple execution units, simplifying the programming of the processor. The full suite of tools and libraries, along with available reference designs, enable designers to efficiently build, debug, profile and optimize their embedded vision systems.
Object Detection Demo with DesignWare EV Family of Vision Processors
In this speed sign detection demo, see how the DesignWare EV vision processors offer high accuracy and performance for embedded vision applications. The EV vision processors are built on a multicore architecture that is optimized for vision applications, and implement a convolutional neural network (CNN) that can operate at more than 1000 GOPS/Watt.
Mike Thompson Sr. Product Marketing Manager, Synopsys
Downloads and Documentation
Optimized for high-performance embedded vision applications
Based on advanced ARCv2 ISA
Delivers 1000 GOPS/W with 5x better power efficiency than GPUs
Fast, accurate object detection with programmable Convolution Neural Network object detection engine
Dual-core and quad-core versions with up to 8 object detection engine Processing Elements
Supports data- and task-level parallelism
Runs broad range of vision algorithm
Works with all host processors for vision offload
High productivity programming tools with OpenCV library and OpenVX runtime and kernels
OpenVX and the OpenVX logo are trademarks of the Khronos Group Inc.