CAREER: Combined Shape- and Intensity-Based Methods for Tracking

职业：基于形状和强度的组合跟踪方法

• 批准号: 0133144
• 负责人: Daniel Freedman
• 金额: $ 35万
• 依托单位: Rensselaer Polytechnic Institute
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2002
• 资助国家: 美国
• 起止时间: 2002-04-01 至 2008-03-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0133144&HistoricalAwards=false
• 关键词: CAREER; Combined; Shape; Intensity; Based

This project focuses on studying general purpose visual tracking systems. A model-based approach will be used, emphasizing the particular class of objects to be tracked. The tracking procedure is to be based on solving partial differential equations describing curve flow. Schemes for performing tracking will be developed based on intensity information, which is contained within the region of the curve in question. Faster, more effective, and more robust trackers will be generated by extending the general framework via generalization of the assumptions used, and through formal analysis of the related PDEs. Also addressed will be the more general problem of combining shape and intensity-based information; as above, the algorithm will also look for intensity patterns that match what it knows about the class of objects to be tracked. The education part of the project includes curriculum development, undergraduate independent studies, and exposure of a large number students to an experimental system designed by the PI.

本项目主要研究通用视觉跟踪系统。 将采用基于模型的方法，强调要跟踪的特定类别的物体。跟踪过程是基于求解描述曲线流的偏微分方程。将根据所讨论的曲线区域内所含的强度信息制定跟踪方案。更快，更有效，更强大的跟踪器将通过扩展的一般框架，通过推广使用的假设，并通过正式的相关偏微分方程的分析。还将解决更一般的问题，结合形状和强度为基础的信息;如上所述，该算法还将寻找强度模式，匹配它所知道的要跟踪的对象的类。该项目的教育部分包括课程开发、本科生独立学习以及让大量学生接触PI设计的实验系统。