Learning Occupancy Function from Point Clouds for Surface Reconstruction
This work addresses the problem of accurate 3D shape reconstruction for computer vision and graphics applications, representing an incremental improvement over existing methods.
The paper tackles surface reconstruction from sparse point clouds by proposing a novel method that adapts Point Convolution Neural Networks (PCNN) to learn occupancy functions, achieving state-of-the-art performance on ShapeNet and demonstrating good generalization on the McGill 3D dataset.
Implicit function based surface reconstruction has been studied for a long time to recover 3D shapes from point clouds sampled from surfaces. Recently, Signed Distance Functions (SDFs) and Occupany Functions are adopted in learning-based shape reconstruction methods as implicit 3D shape representation. This paper proposes a novel method for learning occupancy functions from sparse point clouds and achieves better performance on challenging surface reconstruction tasks. Unlike the previous methods, which predict point occupancy with fully-connected multi-layer networks, we adapt the point cloud deep learning architecture, Point Convolution Neural Network (PCNN), to build our learning model. Specifically, we create a sampling operator and insert it into PCNN to continuously sample the feature space at the points where occupancy states need to be predicted. This method natively obtains point cloud data's geometric nature, and it's invariant to point permutation. Our occupancy function learning can be easily fit into procedures of point cloud up-sampling and surface reconstruction. Our experiments show state-of-the-art performance for reconstructing With ShapeNet dataset and demonstrate this method's well-generalization by testing it with McGill 3D dataset \cite{siddiqi2008retrieving}. Moreover, we find the learned occupancy function is relatively more rotation invariant than previous shape learning methods.