# PointConvFormer: Revenge of the Point-based Convolution

arXiv: https://arxiv.org/abs/2208.02879 · HTML: https://mlsanity.com/paper/2208.02879/

## Key facts

- Cited by 30 papers in the map (data through 14 Oct 2026).
- Builds on 76 of its 107 references that are in the map.
- First posted on arXiv 4 Aug 2022.

## Builds on (most cited first, 30 of 76)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation](https://mlsanity.com/paper/1612.00593/) (2016)
- [PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space](https://mlsanity.com/paper/1706.02413/) (2017)
- [MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications](https://mlsanity.com/paper/1704.04861/) (2017)
- [ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes](https://mlsanity.com/paper/1702.04405/) (2017)
- [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://mlsanity.com/paper/1906.08237/) (2019)
- [CBAM: Convolutional Block Attention Module](https://mlsanity.com/paper/1807.06521/) (2018)
- [Non-local Neural Networks](https://mlsanity.com/paper/1711.07971/) (2017)
- [Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context](https://mlsanity.com/paper/1901.02860/) (2019)
- [PointPillars: Fast Encoders for Object Detection from Point Clouds](https://mlsanity.com/paper/1812.05784/) (2018)
- [Deformable Convolutional Networks](https://mlsanity.com/paper/1703.06211/) (2017)
- [Point Transformer](https://mlsanity.com/paper/2011.00931/) (2020)
- [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)
- [Self-Attention with Relative Position Representations](https://mlsanity.com/paper/1803.02155/) (2018)
- [A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimation](https://mlsanity.com/paper/1512.02134/) (2015)
- [PointCNN: Convolution On X-Transformed Points](https://mlsanity.com/paper/1801.07791/) (2018)
- [3D Semantic Segmentation with Submanifold Sparse Convolutional Networks](https://mlsanity.com/paper/1711.10275/) (2017)
- [Multi-View 3D Object Detection Network for Autonomous Driving](https://mlsanity.com/paper/1611.07759/) (2016)
- [PointConv: Deep Convolutional Networks on 3D Point Clouds](https://mlsanity.com/paper/1811.07246/) (2018)
- [RandLA-Net: Efficient Semantic Segmentation of Large-Scale Point Clouds](https://mlsanity.com/paper/1911.11236/) (2019)
- [Multi-view Convolutional Neural Networks for 3D Shape Recognition](https://mlsanity.com/paper/1505.00880/) (2015)
- [Stand-Alone Self-Attention in Vision Models](https://mlsanity.com/paper/1906.05909/) (2019)
- [DeepGCNs: Can GCNs Go as Deep as CNNs?](https://mlsanity.com/paper/1904.03751/) (2019)
- [Semantic Scene Completion from a Single Depth Image](https://mlsanity.com/paper/1611.08974/) (2016)
- [Searching Efficient 3D Architectures with Sparse Point-Voxel Convolution](https://mlsanity.com/paper/2007.16100/) (2020)

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

- [Point Transformer V3: Simpler, Faster, Stronger](https://mlsanity.com/paper/2312.10035/) (2023)
- [Swin3D: A Pretrained Transformer Backbone for 3D Indoor Scene Understanding](https://mlsanity.com/paper/2304.06906/) (2023)
- [Sonata: Self-Supervised Learning of Reliable Point Representations](https://mlsanity.com/paper/2503.16429/) (2025)
- [Towards Large-scale 3D Representation Learning with Multi-dataset Point Prompt Training](https://mlsanity.com/paper/2308.09718/) (2023)
- [Mask-Attention-Free Transformer for 3D Instance Segmentation](https://mlsanity.com/paper/2309.01692/) (2023)
- [Grounding 3D Object Affordance with Language Instructions, Visual Observations and Interactions](https://mlsanity.com/paper/2504.04744/) (2025)
- [LitePT: Lighter Yet Stronger Point Transformer](https://mlsanity.com/paper/2512.13689/) (2025)
- [KPConvX: Modernizing Kernel Point Convolution with Kernel Attention](https://mlsanity.com/paper/2405.13194/) (2024)
- [3D Open-Vocabulary Panoptic Segmentation with 2D-3D Vision-Language Distillation](https://mlsanity.com/paper/2401.02402/) (2024)
- [Multi-Body Neural Scene Flow](https://mlsanity.com/paper/2310.10301/) (2023)
- [Spherical Frustum Sparse Convolution Network for LiDAR Point Cloud Semantic Segmentation](https://mlsanity.com/paper/2311.17491/) (2023)
- [DiffSF: Diffusion Models for Scene Flow Estimation](https://mlsanity.com/paper/2403.05327/) (2024)
- [FlowMamba: Learning Point Cloud Scene Flow with Global Motion Propagation](https://mlsanity.com/paper/2412.17366/) (2024)
- [Advancements in Point Cloud-Based 3D Defect Detection and Classification for Industrial Systems: A Comprehensive Survey](https://mlsanity.com/paper/2402.12923/) (2024)
- [Object Dynamics Modeling with Hierarchical Point Cloud-based Representations](https://mlsanity.com/paper/2404.06044/) (2024)
- [Self-Supervised 3D Scene Flow Estimation and Motion Prediction using Local Rigidity Prior](https://mlsanity.com/paper/2310.11284/) (2023)
- [PoCo: Point Context Cluster for RGBD Indoor Place Recognition](https://mlsanity.com/paper/2404.02885/) (2024)
- [BiEquiFormer: Bi-Equivariant Representations for Global Point Cloud Registration](https://mlsanity.com/paper/2407.08729/) (2024)
- [CSCPR: Cross-Source-Context Indoor RGB-D Place Recognition](https://mlsanity.com/paper/2407.17457/) (2024)
- [SAMBLE: Shape-Specific Point Cloud Sampling for an Optimal Trade-Off Between Local Detail and Global Uniformity](https://mlsanity.com/paper/2504.19581/) (2025)
- [HyperPointFormer: Multimodal Fusion in 3D Space with Dual-Branch Cross-Attention Transformers](https://mlsanity.com/paper/2505.23206/) (2025)
- [Enhancing point cloud analysis via neighbor aggregation correction based on cross-stage structure correlation](https://mlsanity.com/paper/2506.15160/) (2025)
- [RelFlexformer: Efficient Attention 3D-Transformers for Integrable Relative Positional Encodings](https://mlsanity.com/paper/2605.10706/) (2026)
- [Long Range Pooling for 3D Large-Scale Scene Understanding](https://mlsanity.com/paper/2301.06962/) (2023)
- [Point Neighborhood Embeddings](https://mlsanity.com/paper/2310.02083/) (2023)
- [Bidirectional Knowledge Reconfiguration for Lightweight Point Cloud Analysis](https://mlsanity.com/paper/2310.05125/) (2023)
- [Long-Term 3D Point Tracking By Cost Volume Fusion](https://mlsanity.com/paper/2407.13337/) (2024)
- [PCSTracker: Long-Term Scene Flow Estimation for Point Cloud Sequences](https://mlsanity.com/paper/2603.19762/) (2026)
- [Channel-Level Relation to Attentive Aggregation with Neighborhood-Homogeneity Constraint for Point Cloud Analysis](https://mlsanity.com/paper/2605.02357/) (2026)
- [Rationally Linearized MLPs](https://mlsanity.com/paper/2610.13568/) (2026)

## Abstract (arXiv)

We introduce PointConvFormer, a novel building block for point cloud based deep network architectures. Inspired by generalization theory, PointConvFormer combines ideas from point convolution, where filter weights are only based on relative position, and Transformers which utilize feature-based attention. In PointConvFormer, attention computed from feature difference between points in the neighborhood is used to modify the convolutional weights at each point. Hence, we preserved the invariances from point convolution, whereas attention helps to select relevant points in the neighborhood for convolution. PointConvFormer is suitable for multiple tasks that require details at the point level, such as segmentation and scene flow estimation tasks. We experiment on both tasks with multiple datasets including ScanNet, SemanticKitti, FlyingThings3D and KITTI. Our results show that PointConvFormer offers a better accuracy-speed tradeoff than classic convolutions, regular transformers, and voxelized sparse convolution approaches. Visualizations show that PointConvFormer performs similarly to convolution on flat areas, whereas the neighborhood selection effect is stronger on object boundaries, showing that it has got the best of both worlds.