MSC: Sequential Classification and Detection via Markov Models in Point Clouds of Urban Scenes

从2D和3D图像自动重建场景是3D计算机视觉和图形学中最重要的问题之一。近年来，复杂城市场景的重建引起了人们的极大兴趣。这是因为准确的3D城市模型在城市规划、建筑和考古等各种领域的进一步发展中至关重要。它们对于日常生活中常用的应用程序也非常重要，例如街道地图可视化和导航，以及在电影和建筑行业中。然而，城市场景的三维图像自动重建和分类，是一个其复杂性至今仍面临挑战的问题吗？S研究界。城市模型的三维重建是通过使用激光扫描仪和常规相机等各种设备进行数据采集来实现的。虽然激光扫描仪提供密集、详细和准确的3D点，但它们的速度慢，大大增加了获取成本。该项目由一个多学科团队领导，结合了计算机视觉、数学建模和统计学的专业知识来解决这一限制。这项工作的目标是开发和实现应用于三维点云数据流的实时检测和分类技术。这允许集中获取感兴趣的对象(例如，幕墙等)。从而提高了速度并降低了功耗。它还有助于高级识别过程对城市场景中的对象进行检测和分类。数据的高维性质的降低是通过巧妙地创新选择测量模型来实现的。使用隐马尔可夫模型的新公式来捕捉城市场景的复杂性。实时算法通过最新一代激光距离扫描技术在真实城市环境中获取的点云数据进行测试。在教育方面，该项目为本科生和研究生提供了一个激励的研究环境，涉及数学和计算机科学的交叉学科前沿。它还提供了一个框架，通过开发跨学科课程来创新布鲁克林(一所为少数族裔服务的机构)和亨特学院的课程。

One of the most important problems in 3D computer vision and graphics is the automatic scene reconstruction from 2D and 3D images. Recently, the reconstruction of complex urban scenes has attracted significant interest. This is because accurate 3D city models are paramount in the further development of a variety of fields such as urban planning, architecture, and archeology. They are also very important for applications commonly used in everyday life such as street map visualization and navigation, as well as in the film and construction industries. Automatic 3D image reconstruction and classification of urban scenes, though, is a problem whose complexity still challenges today?s research community. 3D reconstruction of city models is achieved through data acquisition using a variety of devices such as laser scanners and regular cameras. While laser scanners provide dense, detailed and accurate 3D points, they suffer from slow speed which dramatically increases the cost of acquisition.This project, which is led by a multi-disciplinary team, combines expertise from computer vision, mathematical modeling and statistics to address this limitation. The goal of this work is to develop and implement real-time detection and classification techniques applied to streams of 3D point-cloud data. This allows focused acquisition of objects of interest (e.g. facades, etc.) and thus increases speed and reduces power consumption. It also aids high-level recognition processes in detecting and classifying objects in urban scenes. Reduction of the high-dimensional nature of the data is achieved by the clever innovative selection of a measurement model. A new formulation using hidden Markov models is used to capture the complexity of urban scenes. The real-time algorithms are tested on point-cloud data acquired in a real urban setting by the latest generation of laser range scanning technology.On the education front, this project provides a stimulating research environment for both undergraduate and graduate students on the interdisciplinary frontier of mathematics and computer science. It also provides a framework for innovating the curriculum at both Brooklyn, a minority-serving institution, and Hunter Colleges through the development of interdisciplinary courses.