Multi-feature target tracking algorithm of bionic 3D image in virtual reality

The traditional target tracking algorithm is relatively simple, and the tracking effect is not ideal for the interference target. Therefore, in this paper, an target tracking algorithm based on extended Kalman filtering is proposed. First of all, bionic three-dimensional (3D) image feature in virtual reality is extracted; then, the target tracking algorithm based on extended Kalman filtering is used for weight selection and modification of 3D image, to zoom in and out of the window, so as to determine the position of the 3D image features. Experimental results prove that the algorithm has good robustness in the light changes and color interference, and it can meet the real-time requirements of tracking system.