With high-end processing platforms such as the Xilinx Zynq-7000 All Programmable SoC, people want to take full advantage of the processing system and custom peripherals available within the device. an example of this philosophy is a system containing multiple video pipelines in which live video streams are written into memory(input) and memory content is sent out to live video streams(output) while the processor is accessing memory. This application note covers design principles for obtaining high performance from the Zynq-7000 AP SoC memory interfaces, from AXI master interfaces implemented in the programmable logic(PL), and from the ARM Cortex A9 processors.
Video Processing on Xilinx FPGA Tutorial.pdf
The Zybo Z7 can also be used in Xilinx's SDSoC environment, which allows you to design FPGA accelerated programs and video pipelines with ease in an entirely C/C++ environment. For more information on SDSoC, see the Xilinx SDSoC Site. Digilent will be releasing a video capable platform with Linux and OpenCV support in the near future. Note that due to the smaller FPGA in the Zybo Z7-10, only very basic video processing demos will be included with that platform. Digilent recommends the Zybo Z7-20 for those interested in video processing.
Amazon EC2 F1 instances use FPGAs to enable delivery of custom hardware accelerations. F1 instances are easy to program and come with everything you need to develop, simulate, debug, and compile your hardware acceleration code, including an FPGA Developer AMI and supporting hardware level development on the cloud. Using F1 instances to deploy hardware accelerations can be useful in many applications to solve complex science, engineering, and business problems that require high bandwidth, enhanced networking, and very high compute capabilities. Examples of target applications that can benefit from F1 instance acceleration are genomics, search/analytics, image and video processing, network security, electronic design automation (EDA), image and file compression and big data analytics.
Socionext, a leading provider of video processing and distribution products, possesses decades of hardware-based solution expertise essential for the development of a powerful and innovative FPGA-based virtual server platform. By working together with AWS, Socionext offers a hardware-accelerated, cloud-based encoding solution on Amazon EC2 F1 instances that ensures consistent, high-quality video streams while significantly reducing processing time and delivery costs.
The Simulink built-in support for hardware enables students to access the hardware capabilities of the popular credit-card sized, ARM11 based Raspberry Pi computer, from within Simulink environment, and deploy for embedded implementation. Faculty who attend will have a chance to work through lab modules with examples of video and image processing algorithms, from very simple video in/out handling to a more sophisticated processing such as object recognition and edge detection. They will have an opportunity to gain practical hands-on experience in building such high-level examples themselves, and by extension understand the potential for use in the classroom with undergraduate students. 2ff7e9595c
Comments