Virtual video call method and terminals

Abstract

The invention provides a virtual video call method and terminals. The method comprises the steps of collecting a video image of a first terminal user; performing facial recognition on the video image so as to obtain facial expression information; sending the facial expression information to a second terminal which establishes call connection with a first terminal, and enabling the facial expression information to be used for enabling the second terminal to synthesize and display video images according to the facial expression information and a facial image model pre-arranged on the second terminal. According to the method, a facial recognition technology is utilized to extract the facial expression information at a sending end such as the first terminal, and synthesis and restoration of the facial image are achieved at a receiving end such as the second terminal according to the sent facial expression information and a pre-arranged facial image module; due to the fact that the volume of transmitted facial expression data is extremely small, the volume of the data transmitted in the video call process is greatly reduced, video call is smoother, and the influence on the video call caused by network bandwidth limitation or flow limitation is reduced.

Description

technical field [0001] The present invention relates to the field of communication technologies, in particular to a virtual video call method and terminal. Background technique [0002] With the rapid improvement of network broadband and the development and popularization of hardware equipment, the video call market has also entered the fast lane of development. At present, the main method of virtual video calls is to collect images at the sending end, determine the facial area in the image, extract the facial feature information in the facial area, send the extracted facial feature information to the receiving end, and use the The facial feature information reproduces the corresponding user's facial expression. [0003] The current defect is that, since the facial features of each person are different, the data of the extracted facial feature information is still very large, and the above method also needs to reconstruct a specific object facial model according to the faci...

AI Technical Summary

Problems solved by technology

Therefore, it can be seen that the amount of video data transmitted in the prior art is very large, consumes a large amount of data traffic, and can also cause video calls to be unsmooth, and is not suitable for mobile networks with limited bandwidth or occasions with limited traffic, thus seriously hindering video Popularization and promotion of calls

Images