# Learning Point Embeddings from Shape Repositories for Few-Shot Segmentation

arXiv: https://arxiv.org/abs/1910.01269 · HTML: https://mlsanity.com/paper/1910.01269/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 Oct 2019.

## Built on by (most cited first, 5 of 5)

- [Visual Semantic Segmentation Based on Few/Zero-Shot Learning: An Overview](https://mlsanity.com/paper/2211.08352/) (2022)
- [Semi-supervised 3D shape segmentation with multilevel consistency and part substitution](https://mlsanity.com/paper/2204.08824/) (2022)
- [PriFit: Learning to Fit Primitives Improves Few Shot Point Cloud Segmentation](https://mlsanity.com/paper/2112.13942/) (2021)
- [The Neurally-Guided Shape Parser: Grammar-based Labeling of 3D Shape Regions with Approximate Inference](https://mlsanity.com/paper/2106.12026/) (2021)
- [SHRED: 3D Shape Region Decomposition with Learned Local Operations](https://mlsanity.com/paper/2206.03480/) (2022)

## Abstract (arXiv)

User generated 3D shapes in online repositories contain rich information about surfaces, primitives, and their geometric relations, often arranged in a hierarchy. We present a framework for learning representations of 3D shapes that reflect the information present in this meta data and show that it leads to improved generalization for semantic segmentation tasks. Our approach is a point embedding network that generates a vectorial representation of the 3D points such that it reflects the grouping hierarchy and tag data. The main challenge is that the data is noisy and highly variable. To this end, we present a tree-aware metric-learning approach and demonstrate that such learned embeddings offer excellent transfer to semantic segmentation tasks, especially when training data is limited. Our approach reduces the relative error by 10.2% with 8 training examples, by 11.72% with 120 training examples on the ShapeNet semantic segmentation benchmark, in comparison to the network trained from scratch. By utilizing tag data the relative error is reduced by 12.8% with 8 training examples, in comparison to the network trained from scratch. These improvements come at no additional labeling cost as the meta data is freely available.