本课程介绍计算机视觉,包括图像形成的基本原理、相机成像几何、特征检测和匹配、包括立体的多视图几何、运动估计和跟踪以及分类。课程将讲解相关应用程序的基本开发方法,包括在图像中查找已知模型、从立体图像中恢复深度、相机校准、图像稳定、自动对齐(例如全景)、跟踪和动作识别。
本课程侧重于培养学习者的直觉性和数学思维,进而让学习者了解问题理论与实践的差异。所有的算法都在幻灯片中演示。但请记住约吉·贝拉所说的:从理论上讲,理论和差别没有区别,从实践上讲,它是存在区别的。(爱因斯坦说过类似的话)。在此课程中,大多数时候你不需要应用高层次的库函数,只需使用低到中层算法来分析图像和提取结构信息。
猜您喜欢
推荐帖子
Aaron Bobick
After 16 years at Gerogia Tech, on July 1, 2015, I became Dean of the School Engineering & Applied Science at Washington Univeristy in St. Louis.
Teaching
As of Fall, 2011 I had taken over the Introduction to Computer Vision course CS4495. This course is designed for senior undergraduates as well as graduate students who have had no prior experience with Computer Vision. The web sites are listed under my main page so with a little luck they stay for quite a while. One of the more recent version was done in Fall, 2014.
更多开源项目推荐
热门器件
随便看看
热门下载