虚拟数字人主播的表情生成技术研究  

Research on Expression Generation Technology of Virtual Digital Anchor

作　　者：公正 GONG Zheng(Convergent Media Center,Fangzi district,Weifang city,Weifang 261200,China)

机构地区：[1]潍坊市坊子区融媒体中心,山东潍坊261200

出　　处：《电视技术》2024年第8期216-218,223,共4页Video Engineering

摘　　要：深入探讨虚拟数字人主播在电视新闻播报中的表情生成技术。通过精准捕捉、多维度数据处理、逼真映射及高质量渲染等关键技术,虚拟主播能够生动播报新闻,增强观众情感共鸣和参与度。同时,提出了技术改进建议,旨在提升虚拟数字人主播表情的真实度、自然度、多样性及渲染效率,为虚拟主播技术的创新和发展提供参考,进一步推动新闻传播方式的发展。This paper discusses the expression generation technology of virtual digital anchors in television news broadcasting.Through key technologies such as accurate capture,multi-dimensional data processing,realistic mapping and high-quality rendering,virtual anchors can broadcast news vividly and enhance audience emotional resonance and engagement.Technical improvement suggestions are put forward,aiming at improving the authenticity,naturalness,diversity and rendering efficiency of virtual human anchor expression,providing reference for the innovation and development of virtual anchor technology,and further promoting the development of news communication mode.

关 键 词：虚拟数字人主播 电视新闻播报 表情生成技术 

分 类 号：TN931.2[电子电信—信号与信息处理]