Enhacing building facade using mobile LiDAR

对生成3D模型的兴趣是由广泛的应用激发的，例如视频游戏， 虚拟、增强和混合现实、3D车辆导航、监控、可视化、城市规划、 太阳能潜力分析和地理信息系统。近年来，对3D的需求 包含门、窗或阳台的详细建筑物模型已经显著增加。这些需求出现 特别是3D GPS导航系统和在线服务，如Google Earth和Bing Maps。 目前，自动生成类似盒子的3D粗略建筑物模型是可行的。然而，这些结果 建筑模型缺乏详细的建筑立面信息， 需要街道水平的建筑模型。3D详细建筑模型的创建仍然是手动的， 劳动密集型过程。 LiDAR（光探测和测距）技术已被证明对许多地理应用非常有效， 作为3D城市建模和渲染。移动的LiDAR系统在车辆高速行驶时收集3D和视觉数据 公布的速度生成具有足够细节的高精度3D点的能力使移动的 LiDAR是生成详细3D模型的重要数据源。该项目解决具有挑战性的问题， 通过将街道级移动的LiDAR与粗略的建筑物模型集成，创建3D详细建筑物模型 来自Google地球或其他来源。这包括将部分扫描的LiDAR数据与粗配准 模型和细节从LiDAR传输到模型。 详细的建筑物模型可以提高3D地图的精度，也有许多潜在的应用， 建筑信息模型（BIM）和城市规划等。所提出的方法是一种具有成本效益的 该解决方案和自动化流程将显著提高详细3D地图生成的生产力。的 研究结果将对测绘和导航行业、基于位置的服务 提供者，以及整个加拿大的社会福利。

The interest in generating 3D models is motivated by a wide range of applications, such as video games, Virtual, Augmented, and Mixed Reality, 3D vehicle navigation, surveillance, visualization, urban planning, solar potential analysis, and Geographic Information Systems (GIS). In recent years, the demand for 3D detailed building models containing doors, windows or balconies has dramatically increased. These needs arise particularly from 3D GPS navigation systems and on-line services such as Google Earth and Bing Maps. Currently, automatic generation of box-like 3D coarse building models is viable. However, these resultant building models lack detailed building facade information and are not suitable for applications where detailed building models at street level is required. The creation of 3D detailed building models is still a manual and labor-intensive process. LiDAR (Light Detection And Ranging) technology, has proven very effective for many geo-applications such as 3D urban modeling and rendering. Mobile LiDAR systems collect 3D and visual data while vehicles drive at a posted speed. The ability of producing highly accurate 3D points with sufficient detail make the mobile LiDAR a great data source for detailed 3D model generation. This project addresses challenging problems in creating 3D detailed building models by integrating street-level mobile LiDAR with coarse building models from Google Earth or other sources. This includes registration of partially scanned LiDAR data with coarse models and detail transfer from LiDAR to the models. The detailed building models can improve 3D map accuracy and also have many potential applications in building information modeling (BIM) and urban planning etc. The proposed approach is a cost-effective solution and the automated process will significantly improve productivity of detailed 3D map generation. The results will have a substantial economic impact on the mapping and navigation industry, location-based service providers, as well as social benefits to Canada as a whole.