一种新的单张照片三维人脸重建方法

摘要

In this paper, the problem of formation control and obstacle avoidance for multiple mobile robots is discussed. First the formation construction is established, and the formulation of formation control for avoiding the obstacles is given. By using potential function, a formation control scheme is proposed for multiple mobile robots with obstacle avoidance. Under the proposed distributed control scheme, the closed-loop stability and safety are guaranteed theoretically. Simulation results demonstrate the effectiveness of the proposed control strategies.%针对利用反向传播神经网络实现单张照片三维人脸重建方法中速度慢、精度低的缺点,提出了一种改进反向传播神经网络,该方法通过引入混沌变异思想有效解决了APSO算法中的早熟问题,然后以此APSO算法取代传统反向传播神经网络的梯度下降算法,通过对基于传统反向传播神经网络和改进反向传播神经网络的单张照片三维人脸重建方法对比实验,结果表明,采用改进后的反向传播神经网络算法实现的单张照片三维人脸重建,重建的三维人脸更加逼真,并且模型重建的效率和精度也得到了明显提高,重建的时间也大大减少.