Memory-Guided Point Cloud Completion for Dental Reconstruction
This work addresses the challenge of accurate dental reconstruction from incomplete scans, which is important for dental professionals, but it is incremental as it builds on existing encoder-decoder methods with a plug-and-play memory module.
The paper tackles the problem of partial dental point cloud completion, which suffers from large missing regions due to occlusion and limited scanning views, by proposing a retrieval-augmented framework that integrates a prototype memory into encoder-decoder pipelines, resulting in consistent improvements in Chamfer Distance and sharper visual features like cusps and ridges.
Partial dental point clouds often suffer from large missing regions caused by occlusion and limited scanning views, which bias encoder-only global features and force decoders to hallucinate structures. We propose a retrieval-augmented framework for tooth completion that integrates a prototype memory into standard encoder--decoder pipelines. After encoding a partial input into a global descriptor, the model retrieves the nearest manifold prototype from a learnable memory and fuses it with the query feature through confidence-gated weighting before decoding. The memory is optimized end-to-end and self-organizes into reusable tooth-shape prototypes without requiring tooth-position labels, thereby providing structural priors that stabilize missing-region inference and free decoder capacity for detail recovery. The module is plug-and-play and compatible with common completion backbones, while keeping the same training losses. Experiments on a self-processed Teeth3DS benchmark demonstrate consistent improvements in Chamfer Distance, with visualizations showing sharper cusps, ridges, and interproximal transitions. Our approach provides a simple yet effective way to exploit cross-sample regularities for more accurate and faithful dental point-cloud completion.