Abstract | ||
---|---|---|
Large-scale 3D scene has an important application prospect in virtual reality and augmented reality. However, due to the doubling of the amount of data in the 3D reconstruction of large-scale outdoor scene, the index of time becomes a great challenge under the condition of maintaining a certain degree of accuracy. In this paper, we provide a set of methods of aerial photography data acquisition and cluster-based 3D model reconstruction for large-scale scene. Firstly, for the data acquisition end, we adopt two strategies of track pre-planning and feedback-based trajectory planning to meet the requirements of efficient data acquisition. Secondly, we have designed and implemented a distributed 3D reconstruction system to process a large number of aerial photographs, which can quickly and robustly reconstruct large-scale 3D models. Finally, we simulate the crowd behavior based on the PEM model in the reconstructed 3D scene, which has a useful guiding significance for people's daily activities and emergency problems. |
Year | DOI | Venue |
---|---|---|
2019 | 10.1109/ISMAR-Adjunct.2019.00-20 | 2019 IEEE International Symposium on Mixed and Augmented Reality Adjunct (ISMAR-Adjunct) |
Keywords | Field | DocType |
Large-Scale,Distributed modeling,Path planning,Crowd simulation | Motion planning,Computer vision,Aerial photography,Virtual reality,Computer science,Data acquisition,Augmented reality,Crowd simulation,Artificial intelligence,Crowd psychology,3D reconstruction | Conference |
ISBN | Citations | PageRank |
978-1-7281-4766-6 | 0 | 0.34 |
References | Authors | |
9 | 5 |