# Memory-Guided Point Cloud Completion for Dental Reconstruction

arXiv: https://arxiv.org/abs/2512.03598 · HTML: https://mlsanity.com/paper/2512.03598/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 17 references that are in the map.
- First posted on arXiv 3 Dec 2025.

## Builds on (most cited first, 10 of 10)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Retrieval-Augmented Generation for Large Language Models: A Survey](https://mlsanity.com/paper/2312.10997/) (2023)
- [A Point Set Generation Network for 3D Object Reconstruction from a Single Image](https://mlsanity.com/paper/1612.00603/) (2016)
- [PCN: Point Completion Network](https://mlsanity.com/paper/1808.00671/) (2018)
- [PoinTr: Diverse Point Cloud Completion with Geometry-Aware Transformers](https://mlsanity.com/paper/2108.08839/) (2021)
- [ProxyFormer: Proxy Alignment Assisted Point Cloud Completion with Missing Part Sensitive Transformer](https://mlsanity.com/paper/2302.14435/) (2023)
- [SVDFormer: Complementing Point Cloud via Self-view Augmentation and Self-structure Dual-generator](https://mlsanity.com/paper/2307.08492/) (2023)
- [From Mesh Completion to AI Designed Crown](https://mlsanity.com/paper/2501.04914/) (2025)
- [PointSea: Point Cloud Completion via Self-structure Augmentation](https://mlsanity.com/paper/2502.17053/) (2025)
- [Variational Autoencoding of Dental Point Clouds](https://mlsanity.com/paper/2307.10895/) (2023)

## Abstract (arXiv)

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.