3D Point Capsule Networks
This addresses the problem of handling 3D point cloud data for researchers and practitioners in computer vision and robotics, with incremental advancements in existing methods.
The paper tackles processing sparse 3D point clouds by proposing 3D point-capsule networks, an auto-encoder that preserves spatial arrangements, resulting in improvements for tasks like object classification, reconstruction, and part segmentation, and enabling new applications such as part interpolation and replacement.
In this paper, we propose 3D point-capsule networks, an auto-encoder designed to process sparse 3D point clouds while preserving spatial arrangements of the input data. 3D capsule networks arise as a direct consequence of our novel unified 3D auto-encoder formulation. Their dynamic routing scheme and the peculiar 2D latent space deployed by our approach bring in improvements for several common point cloud-related tasks, such as object classification, object reconstruction and part segmentation as substantiated by our extensive evaluations. Moreover, it enables new applications such as part interpolation and replacement.