Skeleton-bridged Point Completion: From Global Inference to Local Adjustment
This work addresses point completion for real-world objects with diverse topologies, offering improved surface details and mesh generation, though it is incremental in its approach.
The paper tackles the problem of completing missing geometries from partial point clouds by proposing a skeleton-bridged network that first predicts 3D skeletons for global structure and then learns displacements for surface reconstruction, outperforming existing methods in qualitative and quantitative experiments across various object categories.
Point completion refers to complete the missing geometries of objects from partial point clouds. Existing works usually estimate the missing shape by decoding a latent feature encoded from the input points. However, real-world objects are usually with diverse topologies and surface details, which a latent feature may fail to represent to recover a clean and complete surface. To this end, we propose a skeleton-bridged point completion network (SK-PCN) for shape completion. Given a partial scan, our method first predicts its 3D skeleton to obtain the global structure, and completes the surface by learning displacements from skeletal points. We decouple the shape completion into structure estimation and surface reconstruction, which eases the learning difficulty and benefits our method to obtain on-surface details. Besides, considering the missing features during encoding input points, SK-PCN adopts a local adjustment strategy that merges the input point cloud to our predictions for surface refinement. Comparing with previous methods, our skeleton-bridged manner better supports point normal estimation to obtain the full surface mesh beyond point clouds. The qualitative and quantitative experiments on both point cloud and mesh completion show that our approach outperforms the existing methods on various object categories.