# SampleNet: Differentiable Point Cloud Sampling

arXiv: https://arxiv.org/abs/1912.03663 · HTML: https://mlsanity.com/paper/1912.03663/

## Key facts

- Cited by 74 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 8 Dec 2019.

## Built on by (most cited first, 30 of 74)

- [Mini-Splatting: Representing Scenes with a Constrained Number of Gaussians](https://mlsanity.com/paper/2403.14166/) (2024)
- [Taming 3DGS: High-Quality Radiance Fields with Limited Resources](https://mlsanity.com/paper/2406.15643/) (2024)
- [Surface Representation for Point Clouds](https://mlsanity.com/paper/2205.05740/) (2022)
- [PnP-DETR: Towards Efficient Visual Analysis with Transformers](https://mlsanity.com/paper/2109.07036/) (2021)
- [SASA: Semantics-Augmented Set Abstraction for Point-based 3D Object Detection](https://mlsanity.com/paper/2201.01976/) (2022)
- [RBGNet: Ray-based Grouping for 3D Object Detection](https://mlsanity.com/paper/2204.02251/) (2022)
- [Attention-based Point Cloud Edge Sampling](https://mlsanity.com/paper/2302.14673/) (2023)
- [FESTA: Flow Estimation via Spatial-Temporal Attention for Scene Point Clouds](https://mlsanity.com/paper/2104.00798/) (2021)
- [Learning Inner-Group Relations on Point Clouds](https://mlsanity.com/paper/2108.12468/) (2021)
- [Correspondence-Free Point Cloud Registration with SO(3)-Equivariant Implicit Shape Representations](https://mlsanity.com/paper/2107.10296/) (2021)
- [An Adjustable Farthest Point Sampling Method for Approximately-sorted Point Cloud Data](https://mlsanity.com/paper/2208.08795/) (2022)
- [3D-QueryIS: A Query-based Framework for 3D Instance Segmentation](https://mlsanity.com/paper/2211.09375/) (2022)
- [Greedy-Based Feature Selection for Efficient LiDAR SLAM](https://mlsanity.com/paper/2103.13090/) (2021)
- [Revisiting Point Cloud Simplification: A Learnable Feature Preserving Approach](https://mlsanity.com/paper/2109.14982/) (2021)
- [PointCA: Evaluating the Robustness of 3D Point Cloud Completion Models Against Adversarial Examples](https://mlsanity.com/paper/2211.12294/) (2022)
- [LighTN: Light-weight Transformer Network for Performance-overhead Tradeoff in Point Cloud Downsampling](https://mlsanity.com/paper/2202.06263/) (2022)
- [PSNet: Fast Data Structuring for Hierarchical Deep Learning on Point Cloud](https://mlsanity.com/paper/2205.14965/) (2022)
- [Task-Aware Sampling Layer for Point-Wise Analysis](https://mlsanity.com/paper/2107.04291/) (2021)
- [Local and Global Point Cloud Reconstruction for 3D Hand Pose Estimation](https://mlsanity.com/paper/2112.06389/) (2021)
- [Meta-Sampler: Almost-Universal yet Task-Oriented Sampling for Point Clouds](https://mlsanity.com/paper/2203.16001/) (2022)
- [A Correct-and-Certify Approach to Self-Supervise Object Pose Estimators via Ensemble Self-Training](https://mlsanity.com/paper/2302.06019/) (2023)
- [ULSR-GS: Ultra Large-scale Surface Reconstruction Gaussian Splatting with Multi-View Geometric Consistency](https://mlsanity.com/paper/2412.01402/) (2024)
- [Boosting Point-BERT by Multi-choice Tokens](https://mlsanity.com/paper/2207.13226/) (2022)
- [MICAS: Multi-grained In-Context Adaptive Sampling for 3D Point Cloud Processing](https://mlsanity.com/paper/2411.16773/) (2024)
- [APP-Net: Auxiliary-point-based Push and Pull Operations for Efficient Point Cloud Classification](https://mlsanity.com/paper/2205.00847/) (2022)
- [Learnable Chamfer Distance for Point Cloud Reconstruction](https://mlsanity.com/paper/2312.16582/) (2023)
- [PointLIE: Locally Invertible Embedding for Point Cloud Sampling and Recovery](https://mlsanity.com/paper/2104.14769/) (2021)
- [Medial Spectral Coordinates for 3D Shape Analysis](https://mlsanity.com/paper/2111.13295/) (2021)
- [PIDS: Joint Point Interaction-Dimension Search for 3D Point Cloud](https://mlsanity.com/paper/2211.15759/) (2022)
- [Edge Aware Learning for 3D Point Cloud](https://mlsanity.com/paper/2309.13472/) (2023)

## Abstract (arXiv)

There is a growing number of tasks that work directly on point clouds. As the size of the point cloud grows, so do the computational demands of these tasks. A possible solution is to sample the point cloud first. Classic sampling approaches, such as farthest point sampling (FPS), do not consider the downstream task. A recent work showed that learning a task-specific sampling can improve results significantly. However, the proposed technique did not deal with the non-differentiability of the sampling operation and offered a workaround instead. We introduce a novel differentiable relaxation for point cloud sampling that approximates sampled points as a mixture of points in the primary input cloud. Our approximation scheme leads to consistently good results on classification and geometry reconstruction applications. We also show that the proposed sampling method can be used as a front to a point cloud registration network. This is a challenging task since sampling must be consistent across two different point clouds for a shared downstream task. In all cases, our approach outperforms existing non-learned and learned sampling alternatives. Our code is publicly available at https://github.com/itailang/SampleNet.