3D Augmented Urban Space Modeling for Smart City

室内外建筑的三维重建是一个越来越重要的研究问题，在全球范围内（例如b谷歌Earth），人们正在大规模地努力恢复城市的室外模型。不幸的是，这些模型在应用中受到限制，它们的真正潜力没有实现。在室内和公用事业（即电力线网络）建模方面也严重缺乏知识和理解，这严重阻碍了推动这些语义丰富的3D空间模型用于工程和商业目的的运动。其中一个应用包括将三维模型扩展到城市信息和通信技术（ICT），以提供实时决策支持和空间感知。这些增强的3D模型作为重要的以模型为中心的接口，可以为有效监测和管理基础设施带来前所未有的机会，从而支持城市的可持续性，提高城市的竞争力，这些城市占全球人口增长的近90%，财富创造的80%，能源消耗总量的60%。在当前的发现提案中，我们提出了一种综合方法来开发具有城市可持续性和智慧城市潜在应用的新型3D框架。通过扩展和推进我们正在进行的三维城市空间建模研究，包括建筑和基础设施建模，我们将开发创新技术，整合三维城市空间模型，并通过增强传递实时信息。目标包括1)创建一种创新的众包建模方法，适用于大规模生成室内模型。2)通过扩充库存数据库，增强现有的悬架式电力网测绘系统，以模拟更广泛的线路类型。3)通过引入原始的3D建筑模型-图像对齐技术和基于四叉树结构的新型分层几何哈希表进行精确定位，开发基于3D模型的室内和室外建筑环境的移动增强现实（MAR）应用。我们的研究项目的宏伟目标是开发和实施一个大型增强城市空间模型框架，该模型包括地上和地下设施、自然物体和人造室内/室外结构。高质量的室内和室外3D城市空间模型的集成，加上来自建筑信息模型（BIM）等系统的数据，将对迈向空间和环境感知社会具有重要意义。最终，我们的目标是消除虚拟空间和物理空间之间的差距，使它们可以被视为同一维度的连续体的一部分，以支持城市可持续性和智慧城市。随着移动计算技术的进步，虚拟空间和物理空间之间的界限不断模糊，空间和上下文感知是未来3D建模和增强研究的趋势。像MAR这样的应用程序可以识别和响应现实世界的内容，并将用户对现实的看法与特定位置的信息结合起来。这些信息可以是简单的文本、图像、多媒体或3D图形的形式。无论是设施管理、城市规划、导航、旅游和勘探，这种增强的信息将为用户提供信息（实时），以便更好地在旅途中做出决策。

3D reconstruction of indoor and outdoor buildings is an increasingly important research problem, with large scale efforts underway to recover outdoor models of cities at a global scale (e.g. Google Earth). Unfortunately these models are limited in applications and their true potential not realized. There is also a severe lack of knowledge and understanding in indoor and utility (i.e. power line network) modeling which significantly hinder movements for pushing forward with these semantically rich 3D spatial models for engineering and commercial purposes. One such applications involve the augmentation of the 3D models to urban Information and Communication Technologies (ICT) for providing real time decision support and spatial awareness. These augmented 3D models serve as important model-centric interfaces that could result in unprecedented opportunities for efficiently monitoring and managing infrastructures for supporting urban sustainability and elevating the competitiveness of cities which account for nearly 90% of global population growth, 80% of wealth creation, and 60% of total energy consumption [1].In this current Discovery proposal we propose a comprehensive approach to developing a novel 3D framework with potential applications toward urban sustainability and smart city. By extending and advancing our on-going research for modeling of 3D urban spaces, including building and infrastructure modeling, we will develop innovative techniques for integrating the 3D urban space models and delivering real-time information through augmentation. The objective include 1) creating an innovative crowd-sourced modeling approach that is suitable for large scale generation of indoor models. 2) Enhancing the existing suspension type power line network mapping system to model a wide range of line types through expanding our inventory database. 3) we will develop 3D model-based mobile augmented reality (MAR) applications for indoor and outdoor building environments by introducing an original 3D building model-to-image alignment technique and a new hierarchical Geometric Hash table based on quadtree structure for performing accurate localization. Our grand goal for this research program is to develop and implement a framework for large scale augmented urban space model that includes above and underground facilities, natural objects, and man-made indoor/outdoor structures. The integration of high quality 3D indoor and outdoor urban space models augmented with data from systems such as Building Information Model (BIM) will be significant for moving towards a spatial and context-aware society. Ultimately we aim to remove the gap between virtual and physical spaces where they can be considered as part of a continuum in the same dimension for supporting urban sustainability and smart city.Spatial and context-awareness is the trend for future 3D modeling and augmentation researches as advancements in mobile computing technologies continue to blur the line between virtual and physical spaces. Applications such as MAR recognize and react to the real world content and combine users’ view of reality with location specific information. Such information could be in the form of simple text, image, multi-media, or 3D graphics. Whether it is for facility management, urban planning, navigation, tourism and exploration, such augmented information will provide users with information (in real time) for better decision making on-the-go.