MedPCFM: Improving Medical Point Cloud Completion by Integrating Point Transformers and Flow Matching

This article introduces PCFM, a flow matching approach for medical point cloud completion that integrates Point Transformer v3 (PTv3) with continuous-time generative modeling. The method is evaluated on the SkullFix, SkullBreak, and Mandibular Defect datasets to assess its performance in anatomical reconstruction tasks.

• PCFM achieves state-of-the-art generative performance across all tested datasets while requiring substantially fewer sampling steps than diffusion models.
• It is competitive with deterministic PTv3 baselines and offers up to a 7x speed-up compared to PVCNN backbones at best operating points.
• Empirical scaling trends show consistent gains with higher point resolution and informative trade-offs across different model scales.

The authors consider this significant because it provides a high-throughput, efficient alternative to diffusion-based methods for medical imaging workflows without sacrificing generative quality.