Deep Shape Representation with Sharp Feature Preservation

We present a novel implicit neural representation to reconstruct CAD models from point clouds in high quality. Our method first extracts sharp edge points from input points by a sharp feature detection network and then recovers implicit surfaces preserving the sharp features of ground-truth models. The sharp feature detection network uses a U-Net structure for feature encoding, and the attention module is introduced to improve the accuracy in CAD models. This detection network is lightweighted and accepts input points with arbitrary numbers. Afterwards, we propose an MLP-based network to train implicit representation from input points with its extracted sharp points. A two-stage training process is proposed, and the loss functions are designed for each stage to ensure that the sharp features of the input points are learned while the surface details are fitted. Comparing our method with other SOTA methods in the ABC dataset, our method is significantly superior to the existing nonlearning and learning 3D reconstruction methods in terms of surface approximation quality and sharp feature preservation. Moreover, we can gain spline representations from learned shapes for CAM as the application of our method.