# Label-Efficient Learning on Point Clouds using Approximate Convex Decompositions

arXiv: https://arxiv.org/abs/2003.13834 · HTML: https://mlsanity.com/paper/2003.13834/

## Key facts

- Cited by 27 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 30 Mar 2020.

## Built on by (most cited first, 27 of 27)

- [PartSLIP: Low-Shot Part Segmentation for 3D Point Clouds via Pretrained Image-Language Models](https://mlsanity.com/paper/2212.01558/) (2022)
- [Unsupervised Point Cloud Representation Learning with Deep Neural Networks: A Survey](https://mlsanity.com/paper/2202.13589/) (2022)
- [Self-Contrastive Learning with Hard Negative Sampling for Self-supervised Point Cloud Learning](https://mlsanity.com/paper/2107.01886/) (2021)
- [PartSLIP++: Enhancing Low-Shot 3D Part Segmentation via Multi-View Instance Segmentation and Maximum Likelihood Estimation](https://mlsanity.com/paper/2312.03015/) (2023)
- [PartDistill: 3D Shape Part Segmentation by Vision-Language Model Distillation](https://mlsanity.com/paper/2312.04016/) (2023)
- [A Survey of Label-Efficient Deep Learning for 3D Point Clouds](https://mlsanity.com/paper/2305.19812/) (2023)
- [ZeroPS: High-quality Cross-modal Knowledge Transfer for Zero-Shot 3D Part Segmentation](https://mlsanity.com/paper/2311.14262/) (2023)
- [Weakly Supervised 3D Point Cloud Segmentation via Multi-Prototype Learning](https://mlsanity.com/paper/2205.03137/) (2022)
- [MvDeCor: Multi-view Dense Correspondence Learning for Fine-grained 3D Segmentation](https://mlsanity.com/paper/2208.08580/) (2022)
- [3x2: 3D Object Part Segmentation by 2D Semantic Correspondences](https://mlsanity.com/paper/2407.09648/) (2024)
- [PARIS3D: Reasoning-based 3D Part Segmentation Using Large Multimodal Model](https://mlsanity.com/paper/2404.03836/) (2024)
- [Semi-supervised 3D shape segmentation with multilevel consistency and part substitution](https://mlsanity.com/paper/2204.08824/) (2022)
- [Representing Shape Collections with Alignment-Aware Linear Models](https://mlsanity.com/paper/2109.01605/) (2021)
- [PriFit: Learning to Fit Primitives Improves Few Shot Point Cloud Segmentation](https://mlsanity.com/paper/2112.13942/) (2021)
- [Point Discriminative Learning for Data-efficient 3D Point Cloud Analysis](https://mlsanity.com/paper/2108.02104/) (2021)
- [Active Self-Training for Weakly Supervised 3D Scene Semantic Segmentation](https://mlsanity.com/paper/2209.07069/) (2022)
- [The Neurally-Guided Shape Parser: Grammar-based Labeling of 3D Shape Regions with Approximate Inference](https://mlsanity.com/paper/2106.12026/) (2021)
- [Novel class discovery meets foundation models for 3D semantic segmentation](https://mlsanity.com/paper/2312.03782/) (2023)
- [Self-Supervised Feature Learning from Partial Point Clouds via Pose Disentanglement](https://mlsanity.com/paper/2201.03018/) (2022)
- [Open-Set Semi-Supervised Learning for 3D Point Cloud Understanding](https://mlsanity.com/paper/2205.01006/) (2022)
- [Semi-supervised segmentation of tooth from 3D Scanned Dental Arches](https://mlsanity.com/paper/2208.05539/) (2022)
- [Learning Convex Decomposition via Feature Fields](https://mlsanity.com/paper/2603.09285/) (2026)
- [SHRED: 3D Shape Region Decomposition with Learned Local Operations](https://mlsanity.com/paper/2206.03480/) (2022)
- [Generalized Robot 3D Vision-Language Model with Fast Rendering and Pre-Training Vision-Language Alignment](https://mlsanity.com/paper/2312.00663/) (2023)
- [Semantic-Free Procedural 3D Shapes Are Surprisingly Good Teachers](https://mlsanity.com/paper/2411.17467/) (2024)
- [Concavity-Induced Distance for Unoriented Point Cloud Decomposition](https://mlsanity.com/paper/2306.11051/) (2023)
- [Random Walks in Self-supervised Learning for Triangular Meshes](https://mlsanity.com/paper/2503.00816/) (2025)

## Abstract (arXiv)

The problems of shape classification and part segmentation from 3D point clouds have garnered increasing attention in the last few years. Both of these problems, however, suffer from relatively small training sets, creating the need for statistically efficient methods to learn 3D shape representations. In this paper, we investigate the use of Approximate Convex Decompositions (ACD) as a self-supervisory signal for label-efficient learning of point cloud representations. We show that using ACD to approximate ground truth segmentation provides excellent self-supervision for learning 3D point cloud representations that are highly effective on downstream tasks. We report improvements over the state-of-the-art for unsupervised representation learning on the ModelNet40 shape classification dataset and significant gains in few-shot part segmentation on the ShapeNetPart dataset.Code available at https://github.com/matheusgadelha/PointCloudLearningACD