mlsanityData to 7 Oct 2026

Parameter is Not All You Need: Starting from Non-Parametric Networks for 3D Point Cloud Analysis

arXiv 2303.08134 ↗First posted 14 Mar 2023cs.CV cs.AIcited by 34

Not on the map. Why?

Abstract (arXiv)

We present a Non-parametric Network for 3D point cloud analysis, Point-NN, which consists of purely non-learnable components: farthest point sampling (FPS), k-nearest neighbors (k-NN), and pooling operations, with trigonometric functions. Surprisingly, it performs well on various 3D tasks, requiring no parameters or training, and even surpasses existing fully trained models. Starting from this basic non-parametric model, we propose two extensions. First, Point-NN can serve as a base architectural framework to construct Parametric Networks by simply inserting linear layers on top. Given the superior non-parametric foundation, the derived Point-PN exhibits a high performance-efficiency trade-off with only a few learnable parameters. Second, Point-NN can be regarded as a plug-and-play module for the already trained 3D models during inference. Point-NN captures the complementary geometric knowledge and enhances existing methods for different 3D benchmarks without re-training. We hope our work may cast a light on the community for understanding 3D point clouds with non-parametric methods. Code is available at https://github.com/ZrrSkywalker/Point-NN.

Key facts. Cited by 34 papers in the map (data through 7 Oct 2026). Builds on 70 of its 100 references that are in the map. First posted on arXiv 14 Mar 2023.

Builds on

70 of 100 references.

References · 70

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  4. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  5. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  6. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
  7. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
  8. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,088
  9. 2021Learning to Prompt for Vision-Language Modelscited by 2,269
  10. 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,220
  11. 2020Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domainscited by 2,170
  12. 2020Point Transformercited by 1,462
38 more
  1. 2018On the Spectral Bias of Neural Networkscited by 1,321
  2. 2019KPConv: Flexible and Deformable Convolution for Point Cloudscited by 1,202
  3. 2021CLIP-Adapter: Better Vision-Language Models with Feature Adapterscited by 1,021
  4. 2019PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detectioncited by 890
  5. 2018PointCNN: Convolution On X-Transformed Pointscited by 764
  6. 2020PCT: Point cloud transformercited by 730
  7. 2016Multi-View 3D Object Detection Network for Autonomous Drivingcited by 721
  8. 2018PointConv: Deep Convolutional Networks on 3D Point Cloudscited by 656
  9. 2015Multi-view Convolutional Neural Networks for 3D Shape Recognitioncited by 564
  10. 2016Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modelingcited by 558
  11. 2017Convolutional Sequence to Sequence Learningcited by 550
  12. 2019Revisiting Point Cloud Classification: A New Benchmark Dataset and Classification Model on Real-World Datacited by 532
  13. 2021PointCLIP: Point Cloud Understanding by CLIPcited by 461
  14. 2022PointNeXt: Revisiting PointNet++ with Improved Training and Scaling Strategiescited by 432
  15. 2021Tip-Adapter: Training-free CLIP-Adapter for Better Vision-Language Modelingcited by 429
  16. 2021Conditional Positional Encodings for Vision Transformerscited by 402
  17. 2017Large-scale Point Cloud Semantic Segmentation with Superpoint Graphscited by 384
  18. 2022Rethinking Network Design and Local Geometry in Point Cloud: A Simple Residual MLP Frameworkcited by 376
  19. 2019Relation-Shape Convolutional Neural Network for Point Cloud Analysiscited by 346
  20. 2019Point-Voxel CNN for Efficient 3D Deep Learningcited by 343
  21. 2016OctNet: Learning Deep 3D Representations at High Resolutionscited by 340
  22. 2021An End-to-End Transformer Model for 3D Object Detectioncited by 329
  23. 2022Point-M2AE: Multi-scale Masked Autoencoders for Hierarchical Point Cloud Pre-trainingcited by 263
  24. 2018SpiderCNN: Deep Learning on Point Sets with Parameterized Convolutional Filterscited by 261
  25. 20196-DOF GraspNet: Variational Grasp Generation for Object Manipulationcited by 237
  26. 2021PAConv: Position Adaptive Convolution with Dynamic Kernel Assembling on Point Cloudscited by 211
  27. 2022Frozen CLIP Models are Efficient Video Learnerscited by 181
  28. 2018Tangent Convolutions for Dense Prediction in 3Dcited by 173
  29. 2021Walk in the Cloud: Learning Curves for Point Clouds Shape Analysiscited by 172
  30. 2022Stratified Transformer for 3D Point Cloud Segmentationcited by 166
  31. 2022CLIP2Point: Transfer CLIP to Point Cloud Classification with Image-Depth Pre-trainingcited by 159
  32. 2023Prompt, Generate, then Cache: Cascade of Foundation Models makes Strong Few-shot Learnerscited by 152
  33. 2021Revisiting Point Cloud Shape Classification with a Simple and Effective Baselinecited by 145
  34. 2022Learning 3D Representations from 2D Pre-trained Models via Image-to-Point Masked Autoencoderscited by 143
  35. 2020MVTN: Multi-View Transformation Network for 3D Shape Recognitioncited by 140
  36. 20183DMV: Joint 3D-Multi-View Prediction for 3D Semantic Scene Segmentationcited by 118
  37. 2019DensePoint: Learning Densely Contextual Representation for Efficient Point Cloud Processingcited by 104
  38. 2022CALIP: Zero-Shot Enhancement of CLIP with Parameter-free Attentioncited by 103

50 most cited of 70 All 70 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 11; 2024: 10; 2025: 9; 2026: 4.

34 papers.

Sort:

By area:

  1. 2023Point-Bind & Point-LLM: Aligning Point Cloud with Multi-modality for 3D Understanding, Generation, and Instruction Followingcited by 192
  2. 2023ImageBind-LLM: Multi-modality Instruction Tuningcited by 142
  3. 2023Not All Features Matter: Enhancing Few-shot CLIP with Adaptive Prior Refinementcited by 84
  4. 2023ViewRefer: Grasp the Multi-view Knowledge for 3D Visual Grounding with GPT and Prototype Guidancecited by 70
  5. 2023Radar-Camera Fusion for Object Detection and Semantic Segmentation in Autonomous Driving: A Comprehensive Reviewcited by 57
  6. 2023Point-PEFT: Parameter-Efficient Fine-Tuning for 3D Pre-trained Modelscited by 42
  7. 2023Revisiting Multimodal Representation in Contrastive Learning: From Patch and Token Embeddings to Finite Discrete Tokenscited by 35
  8. 2023WaterScenes: A Multi-Task 4D Radar-Camera Fusion Dataset and Benchmarks for Autonomous Driving on Water Surfacescited by 34
  9. 2024PartSTAD: 2D-to-3D Part Segmentation Task Adaptationcited by 18
  10. 2024No Time to Train: Empowering Non-Parametric Networks for Few-shot 3D Scene Segmentationcited by 18
  11. 2024Unlearnable 3D Point Clouds: Class-wise Transformation Is All You Needcited by 16
  12. 2024GPSFormer: A Global Perception and Local Structure Fitting-based Transformer for Point Cloud Understandingcited by 12
22 more
  1. 2024Point-GN: A Non-Parametric Network Using Gaussian Positional Encoding for Point Cloud Classificationcited by 8
  2. 2024Point-PRC: A Prompt Learning Based Regulation Framework for Generalizable Point Cloud Analysiscited by 6
  3. 2023Achelous++: Power-Oriented Water-Surface Panoptic Perception Framework on Edge Devices based on Vision-Radar Fusion and Pruning of Heterogeneous Modalitiescited by 4
  4. 20243D Point Cloud Network Pruning: When Some Weights Do not Mattercited by 3
  5. 2025Point-LN: A Lightweight Framework for Efficient Point Cloud Classification Using Non-Parametric Positional Encodingcited by 3
  6. 2025Enhancing 3D Point Cloud Classification with ModelNet-R and Point-SkipNetcited by 3
  7. 2024Efficient and Generic Point Model for Lossless Point Cloud Attribute Compressioncited by 2
  8. 2024Diff-PCC: Diffusion-based Neural Compression for 3D Point Cloudscited by 2
  9. 2024Training-Free Point Cloud Recognition Based on Geometric and Semantic Information Fusioncited by 2
  10. 2025Towards a 3D Transfer-based Black-box Attack via Critical Feature Guidancecited by 2
  11. 2025Enhancing point cloud analysis via neighbor aggregation correction based on cross-stage structure correlationcited by 1
  12. 2025Exploring Semantic Masked Autoencoder for Self-supervised Point Cloud Understandingcited by 1
  13. 2026NPNet: A Non-Parametric Network with Adaptive Gaussian-Fourier Positional Encoding for 3D Classification and Segmentationcited by 1
  14. 2026SLNet: A Super-Lightweight Geometry-Adaptive Network for 3D Point Cloud Recognitioncited by 1
  15. 2023Less is More: Towards Efficient Few-shot 3D Semantic Segmentation via Training-free Networkscited by 0
  16. 2023UNeR3D: Versatile and Scalable 3D RGB Point Cloud Generation from 2D Images in Unsupervised Reconstructioncited by 0
  17. 2025Topology-Aware Modeling for Unsupervised Simulation-to-Reality Point Cloud Recognitioncited by 0
  18. 2025RL-AD-Net: Reinforcement Learning Guided Adaptive Displacement in Latent Space for Refined Point Cloud Completioncited by 0
  19. 2025Learning High-Fidelity Cloth Animation via Skinning-Free Image Transfercited by 0
  20. 2025SegGraph: Leveraging Graphs of SAM Segments for Few-Shot 3D Part Segmentationcited by 0
  21. 2026TSNN: A Non-parametric and Interpretable Framework for Traffic Time Series Forecastingcited by 0
  22. 2026PCT-Prompt: A Prompt-Guided Transformer Framework for Dense Prediction Tasks in Point Cloudscited by 0