Depth-First Search Based 3D Point Cloud Coordinate Reconstruction Algorithm

This paper presents a novel approach for three-dimensional point cloud coordinate reconstruction using a depth-first search algorithm combined with rotational convex hull gift wrapping. The proposed method searches for optimal convex hulls by maximizing both the number of faces and the sum of squared face areas. Experimental results demonstrate that our algorithm achieves superior reconstruction accuracy with an average L2 norm error reduction of 18.7% compared to traditional methods. The algorithm shows particular promise for virtual reality applications, especially for placing interactive virtual objects at corners and edges to facilitate user interaction.