Deep Learning

DeepSDF characterizes the symbolic distance function of shape through a potential encoding and feedforward decoder network. When deep convolutional networks are used directly in 3D space, their time and space complexity will increase dramatically. And more classic and compact surface representations (such as triangular ...