Click here to read the AOMedia announcement
Scalable Video Technology (SVT) is a software-based video coding technology that is highly optimized for Intel® Xeon® Scalable processors and Intel® Xeon® D processors. Using the open source SVT encoder core found on this website, it is possible to spread video encoding processing across multiple Intel® Xeon® processors. With many data centers using Intel® Xeon® processors, a real advantage of processing efficiency can be achieved by using a combination of SVT video encoders and Intel® Xeon® processors. The large core count available on modern processors (e.g. up to 56 cores per dual-socket platform) makes it possible to scale the performance of SVT encoders extremely well as a function of the available computational resources.
SVT Project Goals
The objective of the open source Scalable Video Technology (SVT) project is to provide flexible high-performance software encoder core libraries for media and visual cloud developers. Such libraries will accelerate innovation of new visual cloud services, serving as a starting point for developers to build faster and higher-quality full-feature encoder products. SVT is designed for cloud-native scalability, and it provides outstanding tradeoffs between visual quality and performance, for both VOD and live visual cloud applications.
The SVT encoder architecture is designed specifically for x86 processors, and heavily optimized for Intel® Xeon® Scalable processors in particular, and allows for the encoder core to be split into independently operating threads, each thread processing a different segment of the input video source. These segments are run in parallel on different processor cores, without introducing any loss in fidelity. The SVT architecture provides 3 dimensional parallelism optimization - process based parallelism, picture based parallelism, and segment based parallelism. In addition, SVT architecture provides multi-precision-level features like human visual system (HVS) optimized classification resulting in better rate/quality tradeoffs, and Intel® Xeon® processor-specific performance tuning.
- Visual Mode (Tune 0): Optimized for visual quality
- PSNR/SSIM Mode (Tune 1): Default encoder mode; Optimized to maximize the PSNR/SSIM BDRate performance
- VMAF Mode (Tune 2): Optimized to maximize VMAF BDRate performance
SVT and the Open Visual Cloud
AVS3 video coding is a third generation standard developed by China AVS working group. It is designed to adapt multiple application scenarios, such as UHD TV broadcasting, VR and video surveillance.
SVT- AVS3 is Intel's adoption of the standard, based on Scalable Video Technology (SVT) architecture. the result is an optimal trade-off among coding performance, encoding speed, and visual quality. SVT-AVS3 addresses multidimensional parallelism and perceptual based rate-distortion optimization.