摘要
文章利用基于OpenPose的人体姿态识别算法,将动画或者视频中的人物动作进行处理,实现对角色运动数据的捕获;采用基于运动捕捉数据的关键帧提取算法来获取角色运动的关键帧。实验结果表明,该方法具较高的识别准确率,并且能够提取出对原动画或视频具有概括力的关键帧。
This paper in order to capture the character motion data, the human pose recognition algorithm based on OpenPose is used to process the character movements in the animation or video. The key frame extraction algorithm based on the motion capture data is used to obtain the key frames of the character motion. Experimental results show that the method has high recognition accuracy and can extract key frames that have generalization power to the original animation or video.
作者
张腾
Zhang Teng(School of Software&Internet of Things Engineering,Jiangxi University of Finance and Economics,Nanchang 330000,China)
出处
《无线互联科技》
2020年第3期78-79,共2页
Wireless Internet Technology
基金
江西财经大学资助学生科研课题,项目名称:基于OpenPose的角色动作提取,项目编号:xskt19224。