期刊文献+

数字人生成和驱动技术研究

Research on digital human generation and drive technology
下载PDF
导出
摘要 近年来,随着人工智能技术的不断发展,元宇宙开始受到学者们的广泛关注。数字人是元宇宙中最重要的参与者和创造者,是最具价值和影响力的数字资产,因此如何快速生成和驱动数字人是构建元宇宙的关键问题之一,同时数字人在影视动画领域也有重要的应用。本文首先系统梳理了数字人的概念、发展历程,重点探讨了目前主流的数字人生成和驱动技术。其次,本文提出一种基于照片的数字人生成和实时驱动架构,该架构只需要输入人物照片即可构建写实风格的数字人,并能实现数字人表情的稳定驱动。最后,本文展望了数字人未来应用的趋势和挑战。 In recent years,with the continuous development of artificial intelligence technology,metaverse has started to re⁃ceive extensive attention from scholars.Digital human are the most important participants and creators in metaverse,and the most valuable and influential digital assets.Therefore,how to quickly generate and drive digital human is one of the key issues in building metaverse,while digital human also has important applications in the field of film and animation.This paper first systematically compares the concept and development history of digital human,and focuses on the current mainstream digital human generation and driving techniques.Secondly,this paper proposes a photo⁃based digital human generation and real⁃time driving architecture,which can build a realistic⁃style digital human by simply inputting a photo of the character and can achieve stable driving of the digital human expressions.Finally,this paper looks ahead for both trends and challenges of future applications of digital human.
作者 杨文韬 刘沛卿 佟佳欣 柳杨 Yang Wentao;Liu Peiqing;Tong Jiaxin;Liu Yang(GUANGXI ARTS PUBLISHING HOUSE Co.Ltd.;Beijing University of Posts and Telecommunications)
出处 《现代电影技术》 2023年第6期25-32,共8页 Advanced Motion Picture Technology
关键词 数字人生成 数字人驱动 深度学习 人脸关键点 Generation of Digital Human Driving of Digital Human Deep Learning Face Landmark
  • 相关文献

参考文献2

二级参考文献17

  • 1Efroymson, M.A. "Multiple regression analysis." Mathematical methods for digital computers 1 (1960): 191-203.
  • 2Dalal, Navneet, and Bill Triggs. ~tHistograms of oriented gradients for human detection.~ Computer Vision and Pattern Recognition, 2005. CVPR 2005. IEEE Computer Society Conference on. Vol. 1. IEEE, 2005.
  • 3Chang, Chih-Chung, and Chih-Jen Lin. ~LIBSVM: a library for support vector machines.~ ACM Transactions on Intelligent Systems and Technology (TIST)2.3 (2011): 27.
  • 4Joachims T, Finley T, Yu C N J. Cutting-plane training of structural SVMs[J]. Machine Learning, 2009, 77(1): 27-59.
  • 5Keerthi S S, Lin C J. Asymptotic behaviors of support vector machines with Gaussian kernel[J]. Neural computation, 2003, 15(7): 1667-1689.
  • 6Lin, Hsuan-Tien, and Chih-Jen Lin. ~A study on sigmoid kernels for SVM and the training of non-PSD kernels by SMO-type methods.~ submitted to Neural Computation (2003): 1-32.
  • 7Suykens J A K, Vandewalle J. Least squares support vector machine classifie~[J]. Neural processing lette~, 1999, 9(3): 293-300.
  • 8王洋,张涛,李欢利,刘仰川,王心醉,徐川.应用SFM和多视图面片实现牙模三维重建[J].计算机测量与控制,2013,21(4):1067-1070. 被引量:7
  • 9雷成,吴福朝,胡占义.Kruppa方程与摄像机自标定[J].自动化学报,2001,27(5):621-630. 被引量:58
  • 10毕海芸,郑文俊,曾江源,俞晶星,任治坤.SfM摄影测量方法在活动构造定量研究中的应用[J].地震地质,2017,39(4):656-674. 被引量:31

共引文献10

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部