ITR/AP(CISE): Capturing and Modeling Physics from Images

这个项目的目标是通过分析照片和视频来捕捉真实的世界现象的物理特性。 物体运动的捕获问题已经成为计算机视觉和计算机图形学领域的一个主要研究领域，具有广泛的应用前景。 该项目将尝试设计新的计算机视觉技术，用于捕获和建模不同背景下许多不同对象的动态和物理特性。 如果成功，新技术将能够从单个视频流和其他有限的传感器数据中进行更准确的3D运动估计。 最后，该项目将探索这些研究产品的潜在应用领域。

The objective of this project is to capture physical properties of real world phenomenon by analyzing photographs and video. The problem of capturing motion of objects has emerged as a major research area in the fields of computer vision and computer graphics with a wide range of possible applications. This project will attempt to devise new computer vision techniques for capturing and modeling the dynamics and physical characteristics of a number of differing objects in different contexts. If successful, the new techniques will enable more accurate 3D motion estimation from a single video stream and other limited sensor data. Finally, the project will explore potentially valuable application areas for these research products.