Multi-view point cloud registration with adaptive convergence threshold and its application in 3D model retrieval

摘要

Multi-view point cloud registration is a hot topic in the communities of artificial intelligence and multimedia technology. In this paper, we propose a novel framework to reconstruct 3D models with a multi-view point cloud registration algorithm with adaptive convergence threshold, and apply it to 3D model retrieval subsequently. The iterative closest point (ICP) algorithm is implemented with an adaptive convergence threshold, and further combines with a motion average algorithm for the registration of multi-view point cloud data. After the registration process, the applications are designed for 3D model retrieval. The geometric saliency map is computed based on the vertex curvatures. The test facial triangles are selected to compare with the standard facial triangle. The face and non-face models are then discriminated. The experiments and comparisons demonstrate the effectiveness of the proposed framework.