Next-Generation 4K60 Smart Video Encoder : Integration of Powerful AI Features


The broadcast television industry is undergoing rapid technological evolution, transitioning from HD to 4K and moving from 4G to 5G networks. Videos are transformed into streaming formats like H.264/H.265 via encoders and stored on local or cloud-based storage. However, this has also brought forth new challenges, including the storage and conversion of massive amounts of data, as well as the need for manual tagging and labeling of extensive video content for later editing and playback. With the rapid advancement of AI edge computing chip capabilities and the enhanced ability to convert between videos and text, video encoding and AI are no longer separate domains but gradually merging into an inseparable unity.


The next-generation 4K60 broadcast encoder seamlessly integrates top-tier video compression technology, real-time streaming capabilities, and powerful AI recognition capabilities, addressing the current core requirements:

By combining the NVIDIA Jetson Orin NX with YUAN SC710 capture cards, genuine 4K60P broadcast-grade video capture is achieved. The SC710 series capture cards possess 4:4:4 chroma subsampling, 10-bit depth, and VANC capture capabilities. The integration of NVIDIA GPUDirect RDMA technology reduces overall capture latency to an incredibly low 15 microseconds. Jetson Orin NX also incorporates cutting-edge AV1 compression technology, offering similar compression quality to H.265 but at lower bitrates, significantly enhancing data transfer efficiency, reducing network burden, and substantially cutting storage and transmission costs. Furthermore, YUAN's NexVDO SDK integrates multiple broadcasting industry transmission standards, including RTSP, RTMP, TS, NDI, SRT, DanteAV, and its exclusive SkyLink X ultra-low latency transmission protocol, streamlining the process of capture, encoding, and transmission.

Most importantly, the Jetson Orin NX module features two different computational power modules, offering 70/100 TOPS respectively. On this innovative video encoder, customers can effortlessly implement various AI and generative AI algorithm applications. During the encoding process, AI algorithms can extract information from video content and incorporate it into stream's metadata. This not only assists post-production personnel in rapidly searching for required content segments but also facilitates quick marking of important timestamps. The fusion of AI and 4K video encoding in this manner will significantly redefine the specifications of the next-generation encoder, introducing a unique impact on the traditional broadcast television industry.

System Architecture Diagram

Key Benefits

Resource Optimization & Cost Savings : Through efficient AV1 compression, broadcasting entities can judiciously utilize bandwidth without compromising visual quality, leading to considerable storage space savings.


Rich Streaming Protocol Support : The NexVDO SDK offers support for next-generation broadcasting streaming protocols like NDI, SRT, and DanteAV, allowing easy integration with various broadcasting transmission and reception devices. It also supports CBR ( Constant Bit Rate ) streaming technology, enabling stable control of network transmission rates.


Leading Integration of Generative AI : Pioneering the integration of generative AI algorithms, it simultaneously interprets video content during video encoding. Through SIE ( Stream Information Extraction ), it compresses audio encoding and AI-generated information in synchronization, significantly lowering the threshold for developers.


Product Suggestions

Core Technology in SDK