抽象的な

Pickup of large scale point cloud based on GPU

Ming Huang, Yanmin Wang, Yong Zhang, Xinle Fu

With the rapid development of three-dimensional laser scanning technology, ultra-large fine point cloud data has been gradually become an important data source of threedimensional model. In an interactive computer graphics application system, the interactive pickup of graphics is an important method. However, the traditional pick-up algorithm is limited to a situation that CPU-based ray intersection algorithm can only pick up a small amount of data triangular facets. Because the speed of picking up large scale point cloud data is slow, a GPU-based point cloud picking algorithms was presented to solve the problem. The basic idea of the algorithm is that by spatial transformation to convert the point cloud to screen space, and then, the point was calculated which is the nearest to the mouse click point in screen space. The GPU's parallel computing capabilities were used to achieve spatial transformation and distance comparison by Computer Shader in this algorithm. So the speed of the pickup has been increased. The experimental results show that compared with the CPU, the pickup method based on GPU parallel computing has greater speed advantage. Especially for the point cloud over 10 million points, the speed of the pickup has been increased 2-3 times faster. This use of GPU parallel computing capabilities have significant advantages in terms of handling large scale data volumes.