Video and Imaging
Accelerating Live Video Workloads
Introducing the Xilinx Real-Time Video Server appliance reference architectures. Powered by Alveo™ accelerator cards, Xilinx delivers breakthrough live video streaming performance at the lowest cost per channel for significant TCO savings over fixed-architecture approaches, without altering existing infrastructure.
What Xilinx Video & Imaging solutions can do for you
Support ever-changing needs of video and imaging applications, without sacrificing price, performance, or power.
Full library of video &
imaging solutions
Best Performance
per bit
Lowest cost per
image/stream
Lowest power per
image/stream
Solution Advantages
Xilinx FPGAs running on the HEVC codec offers 20X faster encode speed vs. CPUs and 40% lower bits vs. GPUs at the same visual quality. This enables companies to lower bandwidth consumption by 30% and saving >20 million dollars annually for 100K streams.
FFmpeg Video Support
- A complete, cross-platform solution to record, convert and stream audio and video
- The free and open-source project used by developers and software workflows
Adaptive Bit Rate Video Transcoding
- Real-time H.264 to either HEVC or VP9 ABR
- Highest Quality live encoding (equal to x265 slow preset)
- Easy programming with FFmpeg integration
Alveo Acceleration Cards
- Accelerates high-quality H.264, HEVC, VP9, and Perseus+ video encoders
- Accelerates image transcoding with JPEG, Lepton and WebP encoders
Applications and Libraries
Xilinx and its ecosystem partners offer a breadth of video and imaging solutions for customers to choose from. Using these solutions, customers can easily accelerate and deploy their applications without requiring any FPGA design experience.
See all Apps & Libraries
Featured Hardware
Xilinx and its partners provide customers with a range of standard PCIe form-factor accelerator cards using Xilinx Virtex UltraScale+ FPGAs and Zynq UltraScale+ EV devices.
Alveo
Adaptable PCIe accelerator cards for all video and imaging transcoding workloads.
Advantech
The VEGA 4000 Series of FPGA-based PCI Express cards are ideal for accelerating machine learning, data analytics and live video processing applications both in appliances and in scale-out data center servers.
Aupera
The Aupera™ Aup260x distributed video processing system is designed to deliver more performance at a fraction of the energy cost and rack space of a traditional x86 solution.
Case Study
Video streaming will account for over 80% of internet traffic by 20201. Supporting the streaming of social media, gaming, marketing, and broadcasting is putting significant stress on data center infrastructure. Service providers, like Twitch, are transitioning to hardware acceleration and FPGA adaptable computing to simplify infrastructure and lower costs.
Read Twitch Case Study>
Xilinx Video Solution Stack
Xilinx has integrated key video transcoding plug-ins into FFmpeg, enabling simple hardware offloading of compute-intensive workloads to a Xilinx Alveo accelerator card.
