# Number-Adaptive Prototype Learning for 3D Point Cloud Semantic Segmentation

arXiv: https://arxiv.org/abs/2210.09948 · HTML: https://mlsanity.com/paper/2210.09948/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 14 of its 17 references that are in the map.
- First posted on arXiv 18 Oct 2022.

## Builds on (most cited first, 14 of 14)

- [End-to-End Object Detection with Transformers](https://mlsanity.com/paper/2005.12872/) (2020)
- [PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation](https://mlsanity.com/paper/1612.00593/) (2016)
- [4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks](https://mlsanity.com/paper/1904.08755/) (2019)
- [KPConv: Flexible and Deformable Convolution for Point Clouds](https://mlsanity.com/paper/1904.08889/) (2019)
- [Per-Pixel Classification is Not All You Need for Semantic Segmentation](https://mlsanity.com/paper/2107.06278/) (2021)
- [Segmenter: Transformer for Semantic Segmentation](https://mlsanity.com/paper/2105.05633/) (2021)
- [RandLA-Net: Efficient Semantic Segmentation of Large-Scale Point Clouds](https://mlsanity.com/paper/1911.11236/) (2019)
- [Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR Segmentation](https://mlsanity.com/paper/2011.10033/) (2020)
- [SalsaNext: Fast, Uncertainty-aware Semantic Segmentation of LiDAR Point Clouds for Autonomous Driving](https://mlsanity.com/paper/2003.03653/) (2020)
- [SqueezeSegV3: Spatially-Adaptive Convolution for Efficient Point-Cloud Segmentation](https://mlsanity.com/paper/2004.01803/) (2020)
- [Rethinking Semantic Segmentation: A Prototype View](https://mlsanity.com/paper/2203.15102/) (2022)
- [MotionNet: Joint Perception and Motion Prediction for Autonomous Driving Based on Bird's Eye View Maps](https://mlsanity.com/paper/2003.06754/) (2020)
- [SuMa++: Efficient LiDAR-based Semantic SLAM](https://mlsanity.com/paper/2105.11320/) (2021)
- [3D Point Cloud Processing and Learning for Autonomous Driving](https://mlsanity.com/paper/2003.00601/) (2020)

## Built on by (most cited first, 7 of 7)

- [MotionTrack: End-to-End Transformer-based Multi-Object Tracing with LiDAR-Camera Fusion](https://mlsanity.com/paper/2306.17000/) (2023)
- [3D Part Segmentation via Geometric Aggregation of 2D Visual Features](https://mlsanity.com/paper/2412.04247/) (2024)
- [Deep Learning on 3D Semantic Segmentation: A Detailed Review](https://mlsanity.com/paper/2411.02104/) (2024)
- [Advancing Chart Question Answering with Robust Chart Component Recognition](https://mlsanity.com/paper/2407.21038/) (2024)
- [Image-Guided Outdoor LiDAR Perception Quality Assessment for Autonomous Driving](https://mlsanity.com/paper/2406.17265/) (2024)
- [Interpretable Affordance Detection on 3D Point Clouds with Probabilistic Prototypes](https://mlsanity.com/paper/2504.18355/) (2025)
- [Advancing Weakly-Supervised Change Detection in Satellite Images via Adversarial Class Prompting](https://mlsanity.com/paper/2508.17186/) (2025)

## Abstract (arXiv)

3D point cloud semantic segmentation is one of the fundamental tasks for 3D scene understanding and has been widely used in the metaverse applications. Many recent 3D semantic segmentation methods learn a single prototype (classifier weights) for each semantic class, and classify 3D points according to their nearest prototype. However, learning only one prototype for each class limits the model's ability to describe the high variance patterns within a class. Instead of learning a single prototype for each class, in this paper, we propose to use an adaptive number of prototypes to dynamically describe the different point patterns within a semantic class. With the powerful capability of vision transformer, we design a Number-Adaptive Prototype Learning (NAPL) model for point cloud semantic segmentation. To train our NAPL model, we propose a simple yet effective prototype dropout training strategy, which enables our model to adaptively produce prototypes for each class. The experimental results on SemanticKITTI dataset demonstrate that our method achieves 2.3% mIoU improvement over the baseline model based on the point-wise classification paradigm.