EC-Net: an Edge-aware Point set Consolidation Network
This addresses the challenge of accurate 3D surface reconstruction from scans for applications like computer graphics and robotics, though it is an incremental advance over existing neural network approaches.
The paper tackles the problem of consolidating sparse, irregular, and noisy 3D point clouds by introducing EC-Net, the first deep learning-based edge-aware technique. It demonstrates improved 3D reconstructions by outperforming state-of-the-art methods on both synthetic and real point clouds.
Point clouds obtained from 3D scans are typically sparse, irregular, and noisy, and required to be consolidated. In this paper, we present the first deep learning based edge-aware technique to facilitate the consolidation of point clouds. We design our network to process points grouped in local patches, and train it to learn and help consolidate points, deliberately for edges. To achieve this, we formulate a regression component to simultaneously recover 3D point coordinates and point-to-edge distances from upsampled features, and an edge-aware joint loss function to directly minimize distances from output points to 3D meshes and to edges. Compared with previous neural network based works, our consolidation is edge-aware. During the synthesis, our network can attend to the detected sharp edges and enable more accurate 3D reconstructions. Also, we trained our network on virtual scanned point clouds, demonstrated the performance of our method on both synthetic and real point clouds, presented various surface reconstruction results, and showed how our method outperforms the state-of-the-arts.