Method and system for processing split online video on the basis of depth sensor

Abstract

The invention discloses a method and a system for processing split online video on the basis of a depth sensor. The method includes: firstly, extracting the depth sensor features on the basis of a video frame and a depth image corresponding to the same, and subjecting the features to video frame foreground-background splitting to obtain binary images; secondly, subjecting foreground holes of the binary split images to detection and filling to obtain binary images with foreground holes filled; thirdly, subjecting the binary images with the foreground holes filled to border optimization to obtain optimized binary images; and fourthly, fusing the optimized binary images to a virtual background and the video frame to generate virtual-real fused images. The problems that the depth-sensor-based online video splitting easily errs at discontinuous depth positions and the prior art loses balance between accuracy and timeliness are solved. A depth-sensor-based online video post-splitting processing method and a virtual-real fusion system meeting the timeliness requirement and having high quality are provided.

Description

technical field [0001] The present invention relates to the fields of video content analysis, image processing and computer vision, in particular to a processing method and system for online video segmentation based on a depth sensor. Background technique [0002] With the development of ubiquitous computing technology, video coding technology and broadband network technology, it has become a new hot spot in the 21st century to realize common communication and communication between people in different places through remote video on the Internet, and it shows a broad application prospect. In addition to the application scope of traditional administrative meetings and office meetings, the application range of remote video interaction has been extended to telemedicine, distance education, distance business meetings and legal fields. In recent years, remote video interaction has gradually developed towards providing an immersive experience. The purpose is to make participants fe...

AI Technical Summary

Problems solved by technology

[0007] The purpose of the present invention is to solve the problem that the online video segmentation based on the depth sensor is easy to make mistakes at the boundary, and the existing technology has the problem of taking one thing and the other in terms of accuracy and real-time performance, so as to provide a high-quality video that meets the real-time requirements. Depth sensor-based online video segmentation post-processing method and a set of virtual-real fusion system

Images