3D target tracking and shape reconstruction in clutter using Gaussian process and point completion network

Existing problems for 3D extended target tracking and contour reconstruction under clutter environment are thoroughly investigated in this article. Due to the sparsity of available point cloud data interference from clutter, shape completion method, such as Point Completion Network (PCN), cannot reconstruct directly. Moreover, traditional Gaussian process (GP) model suffered computing overhead handle irregular non-convex shapes. Here, a two-stage algorithm is proposed, which firstly combines GP measurement with probability association filter jointly estimate kinematics basis points, contain partial information. Afterwards, method input produced points deep learning-based PCN complete geometry contour. The effectiveness proposed verified by both simple complex geometric simulation results show that obtains accurate kinematic state estimation produces