Non-orthogonal multiple access(NOMA) is proved to be useful to satisfy the requirements of beyond 5th generation such as massive multi-user connection. Here we
目的针对传统非刚性3维模型的对应关系计算方法需要模型间真实对应关系监督的缺点,提出一种自监督深度残差函数映射网络(self-supervised deep residual functional maps network,SSDRFMN)。方法首先将局部坐标系与直方图结合以计算3维模型的特征描述符,即方向直方图签名(signature of histograms of orientations,S