China Unicom leads the successful establishment of AI-based real-time super-resolution service demand standard project

Jiwei.com reported on May 15 that the 16th Study Group of the International Telecommunication Union (ITU-T SG16) held a plenary meeting recently. More than 100 representatives from China, the United States, Germany, Brazil, Spain, Russia, Switzerland, Canada, the United Kingdom, Japan, and South Korea participated in the two-week online meeting. The ITU-T F.AI-RSRSreqs:Requirements for real-time super resolution service based on artificial intelligence (Real-time super resolution service requirements based on AI) international standard project proposal led by China Unicom was officially approved.

It is reported that real-time super-resolution technology is a combination of AI technology and ultra-high-definition technology. With the vigorous development of video services, users' requirements for the resolution and quality of real-time video services are also increasing. Taking the monitoring service as an example, the quality of user video services will be reduced due to problems such as insufficient acquisition capabilities of video monitoring terminals, image quality damage caused by network transmission, and low resolution. For low-resolution videos, super-resolution technology based on artificial intelligence can maximize video resolution, enhance picture details, and smooth picture artifacts and noise while ensuring video detail texture. On the basis of maintaining the real-time nature of video services, the picture quality is significantly improved. In the future, it will have broad application prospects in remote sensing, medical imaging, security monitoring, video live broadcast and other fields.

This standard will conduct in-depth research on the overall process, network architecture, model training, video stream synchronization and other modules of AI-based real-time super-resolution services, and promote the close integration of ultra-high-definition video and AI technology, which is of great significance to the development of ultra-high-definition services.