Scalable AI-Assisted 3D City Model Reconstruction from Multiple Sourced Systems

研究的主要目标是开发一个可靠和实用的人在回路框架，用于多源系统的融合，并创建有用的3D城市表示。目前，操作系统中涉及大量的手工工作，并且仅在非常局部的范围内可用，因为这些实践的劳动密集型性质限制了可扩展性。该项目旨在利用人在回路方法来收集开发流程所需的知识，这些流程可以节省运营成本，覆盖更大的区域/城市/国家，允许更好的时间和资源管理，更快的调查（例如，通过飞行/街头无人机），具有高精度，可用于后续检查和测量。它将培养一名具有以人为本，深度学习，程序建模和3D几何处理背景的博士研究员。

The key objective of the research is to develop a reliable and practical human-in-the-loop framework for the fusion of multi-sourced system and create useful 3D City representation. At the moment, there is a lot of manual work involved in OS and are only made available at a very local scale because the labourintensive nature of the practices limits scalability. The project aims to leverage the human-in-the-loop approach to gather the knowledge required to develop processes that could save operational costs, handlea much larger region/city/nations, allow better time and resource management, faster surveying (e.g., via flying/street drones) with high accuracy for subsequent inspection and measurement. It will train a PhD researcher with the right background in humancentredness, deep learning, procedure modelling and 3D geometry processing.