Representing 3D shapes based on implicit surface functions learned from RBF neural networks.

•We propose a 3D shape representation method based on neural network classifier.•The combination of radial base functions can implicitly represent complex shapes.•The use of neural network can represent the shape with 3 classes of points.•We conduct extensive experiments on medical and non-medical data.•Our method can accurately and memory-efficiently represent shapes.•We introduced a new prostate dataset.