ZTE's video algorithm wins the 2019 DAVIS Semi-supervised Challenge championship at IEEE CVPR 2019

Date:2019-06-21 ZTE Click:310

21 June 2019, Shenzhen, China - ZTE Corporation (0763.HK / 000063.SZ), a major international provider of telecommunications, enterprise and consumer technology solutions for the Mobile Internet, today announced that it has won the DAVIS Semi-supervised Challenge championship at the IEEE CVPR ( Computer Vision and Pattern Recognition ) in Long Beach of California, the U.S., ranking first in technical indicators. Moreover, ZTE ranked second in the decoding speed and top 10 in terms of Peak Signal to Noise Ratio(PSNR) in the Challenge on Learned Image Compression. The achievements indicate that ZTE’s video processing technology has been in the first class of the industry.
ZTE participated in the competition by means of two key leading technologies, video coding and video segmentation, which are the basic technologies of 5G big video services.
Used on cloud, edge and devices, video coding is a key technology of 5G + MEC industry applications. For strong interaction scenarios, the video coding technology can customize low-delay coding algorithms.
Moreover, video coding can support the development of new industry services, such as remote control in heavy industry enterprises and ports, video sorting in smart factories, and drone patrolling.
Over 80% scenarios of ZTE's 5G industrial application solutions provide video backhaul. With current 5G + MEC providing shunting, video coding will fully leverage the low latency capability of end-to-end 5G networks.
As a basic technology of AR, video segmentation can track motions and replace background by retrieving objects from videos. Video segmentation is a key technology for target recognition and intelligent video analysis. It has been put into commercial use in ZTE's Blade series smartphones and video conferencing products. In the future , video segmentation will be applied in 5G + AR video services in the fields of intelligent manufacturing, Internet of Vehicles and health care.