# Decoupled Local Aggregation for Point Cloud Learning

arXiv: https://arxiv.org/abs/2308.16532 · HTML: https://mlsanity.com/paper/2308.16532/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- Builds on 43 of its 50 references that are in the map.
- First posted on arXiv 31 Aug 2023.

## Builds on (most cited first, 30 of 43)

- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift](https://mlsanity.com/paper/1502.03167/) (2015)
- [Rethinking the Inception Architecture for Computer Vision](https://mlsanity.com/paper/1512.00567/) (2015)
- [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)
- [ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes](https://mlsanity.com/paper/1702.04405/) (2017)
- [ShapeNet: An Information-Rich 3D Model Repository](https://mlsanity.com/paper/1512.03012/) (2015)
- [Xception: Deep Learning with Depthwise Separable Convolutions](https://mlsanity.com/paper/1610.02357/) (2016)
- [Dynamic Graph CNN for Learning on Point Clouds](https://mlsanity.com/paper/1801.07829/) (2018)
- [Point Transformer](https://mlsanity.com/paper/2011.00931/) (2020)
- [Deep Sets](https://mlsanity.com/paper/1703.06114/) (2017)
- [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)
- [Deep Networks with Stochastic Depth](https://mlsanity.com/paper/1603.09382/) (2016)
- [PointCNN: Convolution On X-Transformed Points](https://mlsanity.com/paper/1801.07791/) (2018)
- [Network In Network](https://mlsanity.com/paper/1312.4400/) (2013)
- [Multi-view Convolutional Neural Networks for 3D Shape Recognition](https://mlsanity.com/paper/1505.00880/) (2015)
- [Revisiting Point Cloud Classification: A New Benchmark Dataset and Classification Model on Real-World Data](https://mlsanity.com/paper/1908.04616/) (2019)
- [DeepGCNs: Can GCNs Go as Deep as CNNs?](https://mlsanity.com/paper/1904.03751/) (2019)
- [MetaFormer Is Actually What You Need for Vision](https://mlsanity.com/paper/2111.11418/) (2021)
- [Regularizing Neural Networks by Penalizing Confident Output Distributions](https://mlsanity.com/paper/1701.06548/) (2017)
- [PointNeXt: Revisiting PointNet++ with Improved Training and Scaling Strategies](https://mlsanity.com/paper/2206.04670/) (2022)
- [Point Transformer V2: Grouped Vector Attention and Partition-based Pooling](https://mlsanity.com/paper/2210.05666/) (2022)
- [Rethinking Network Design and Local Geometry in Point Cloud: A Simple Residual MLP Framework](https://mlsanity.com/paper/2202.07123/) (2022)
- [Relation-Shape Convolutional Neural Network for Point Cloud Analysis](https://mlsanity.com/paper/1904.07601/) (2019)
- [OctNet: Learning Deep 3D Representations at High Resolutions](https://mlsanity.com/paper/1611.05009/) (2016)
- [PAConv: Position Adaptive Convolution with Dynamic Kernel Assembling on Point Clouds](https://mlsanity.com/paper/2103.14635/) (2021)
- [Walk in the Cloud: Learning Curves for Point Clouds Shape Analysis](https://mlsanity.com/paper/2105.01288/) (2021)
- [Stratified Transformer for 3D Point Cloud Segmentation](https://mlsanity.com/paper/2203.14508/) (2022)
- [Revisiting Point Cloud Shape Classification with a Simple and Effective Baseline](https://mlsanity.com/paper/2106.05304/) (2021)

## Built on by (most cited first, 12 of 12)

- [Point Cloud Mamba: Point Cloud Learning via State Space Model](https://mlsanity.com/paper/2403.00762/) (2024)
- [GPSFormer: A Global Perception and Local Structure Fitting-based Transformer for Point Cloud Understanding](https://mlsanity.com/paper/2407.13519/) (2024)
- [MAP: Unleashing Hybrid Mamba-Transformer Vision Backbone's Potential with Masked Autoregressive Pretraining](https://mlsanity.com/paper/2410.00871/) (2024)
- [X-3D: Explicit 3D Structure Modeling for Point Cloud Recognition](https://mlsanity.com/paper/2404.15010/) (2024)
- [UniPre3D: Unified Pre-training of 3D Point Cloud Models with Cross-Modal Gaussian Splatting](https://mlsanity.com/paper/2506.09952/) (2025)
- [Mitigating Ambiguities in 3D Classification with Gaussian Splatting](https://mlsanity.com/paper/2503.08352/) (2025)
- [MS-DGCNN++: Multi-Scale Dynamic Graph Convolution with Scale-Dependent Normalization for Robust LiDAR Tree Species Classification](https://mlsanity.com/paper/2507.12602/) (2025)
- [PointCSP: Cross-Sample Semantic Propagation and Stability Preservation in Self-Supervised Point Cloud Learning](https://mlsanity.com/paper/2605.01759/) (2026)
- [Unlocking Hidden Potential in Point Cloud Networks with Attention-Guided Grouping-Feature Coordination](https://mlsanity.com/paper/2509.16639/) (2025)
- [3D Gaussian Point Encoders](https://mlsanity.com/paper/2511.04797/) (2025)
- [LIDARLearn: A Unified Deep Learning Library for 3D Point Cloud Classification, Segmentation, and Self-Supervised Representation Learning](https://mlsanity.com/paper/2604.10780/) (2026)
- [Channel-Level Relation to Attentive Aggregation with Neighborhood-Homogeneity Constraint for Point Cloud Analysis](https://mlsanity.com/paper/2605.02357/) (2026)

## Abstract (arXiv)

The unstructured nature of point clouds demands that local aggregation be adaptive to different local structures. Previous methods meet this by explicitly embedding spatial relations into each aggregation process. Although this coupled approach has been shown effective in generating clear semantics, aggregation can be greatly slowed down due to repeated relation learning and redundant computation to mix directional and point features. In this work, we propose to decouple the explicit modelling of spatial relations from local aggregation. We theoretically prove that basic neighbor pooling operations can too function without loss of clarity in feature fusion, so long as essential spatial information has been encoded in point features. As an instantiation of decoupled local aggregation, we present DeLA, a lightweight point network, where in each learning stage relative spatial encodings are first formed, and only pointwise convolutions plus edge max-pooling are used for local aggregation then. Further, a regularization term is employed to reduce potential ambiguity through the prediction of relative coordinates. Conceptually simple though, experimental results on five classic benchmarks demonstrate that DeLA achieves state-of-the-art performance with reduced or comparable latency. Specifically, DeLA achieves over 90% overall accuracy on ScanObjectNN and 74% mIoU on S3DIS Area 5. Our code is available at https://github.com/Matrix-ASC/DeLA .