基于光流导向的流场可视化算法及视频水彩风格化应用研究  

Research on flow visualization algorithm based on optical flow guidance and application of video watercolor stylization

作　　者：赵杨 Zhao Yang(Animation Department,Yunnan Normal University,Kunming Yunnan,650500)

机构地区：[1]云南师范大学传媒学院动画系,云南昆明650500

出　　处：《电子测试》2022年第11期43-46,共4页Electronic Test

摘　　要：流场可视化是科学可视化一个非常热门的子领域,同时也是是图像风格化绘制的重要组成部分。本文主要探讨基于光流导向的二维流场可视化算法的设计与实现,并将该项技术拓展到对视频进行水彩风格化渲染方面。本文的创新点在于结合物理方法模拟以及图像处理技术对视频进行水彩风格化渲染,并进一步提出采用GPU对整个算法进行硬件加速处理,实验取得了较为良好的合成效果。Flow visualization is a very popular sub field of scientific visualization,and it is also an important part of image stylized rendering.This paper mainly discusses the design and implementation of two-dimensional flow visualization algorithm based on optical flow guidance,and extends this technology to watercolor style rendering of video.The innovation of this paper is to combine the physical method simulation and image processing technology to render the video watercolor style,and further propose to use GPU to speed up the hardware processing of the whole algorithm.The experiment has achieved a good synthesis effect.

关 键 词：流场可视化 流体动力学 水彩风格 非真实感绘制 GPU 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]