PriFit: Learning to Fit Primitives Improves Few Shot Point Cloud Segmentation
This addresses label-efficient segmentation for 3D point clouds, offering a semi-supervised approach that is incremental in combining existing techniques.
The paper tackles few-shot 3D point cloud segmentation by proposing PriFit, which learns point representations through geometric primitive fitting, and reports that it outperforms state-of-the-art methods on ShapeNet and PartNet benchmarks.
We present PriFit, a semi-supervised approach for label-efficient learning of 3D point cloud segmentation networks. PriFit combines geometric primitive fitting with point-based representation learning. Its key idea is to learn point representations whose clustering reveals shape regions that can be approximated well by basic geometric primitives, such as cuboids and ellipsoids. The learned point representations can then be re-used in existing network architectures for 3D point cloud segmentation, and improves their performance in the few-shot setting. According to our experiments on the widely used ShapeNet and PartNet benchmarks, PriFit outperforms several state-of-the-art methods in this setting, suggesting that decomposability into primitives is a useful prior for learning representations predictive of semantic parts. We present a number of ablative experiments varying the choice of geometric primitives and downstream tasks to demonstrate the effectiveness of the method.