# FPNN: Field Probing Neural Networks for 3D Data

arXiv: https://arxiv.org/abs/1605.06240 · HTML: https://mlsanity.com/paper/1605.06240/

## Key facts

- Cited by 25 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 May 2016.

## Built on by (most cited first, 25 of 25)

- [Walk in the Cloud: Learning Curves for Point Clouds Shape Analysis](https://mlsanity.com/paper/2105.01288/) (2021)
- [3DILG: Irregular Latent Grids for 3D Generative Modeling](https://mlsanity.com/paper/2205.13914/) (2022)
- [Semantics for Robotic Mapping, Perception and Interaction: A Survey](https://mlsanity.com/paper/2101.00443/) (2021)
- [RIConv++: Effective Rotation Invariant Convolutions for 3D Point Clouds Deep Learning](https://mlsanity.com/paper/2202.13094/) (2022)
- [TopoLogic: An Interpretable Pipeline for Lane Topology Reasoning on Driving Scenes](https://mlsanity.com/paper/2405.14747/) (2024)
- [Deep Learning-based 3D Point Cloud Classification: A Systematic Survey and Outlook](https://mlsanity.com/paper/2311.02608/) (2023)
- [Towards Efficient Graph Convolutional Networks for Point Cloud Handling](https://mlsanity.com/paper/2104.05706/) (2021)
- [Surrogate Modeling of Car Drag Coefficient with Depth and Normal Renderings](https://mlsanity.com/paper/2306.06110/) (2023)
- [Laplacian2Mesh: Laplacian-Based Mesh Understanding](https://mlsanity.com/paper/2202.00307/) (2022)
- [TopoPoint: Enhance Topology Reasoning via Endpoint Detection in Autonomous Driving](https://mlsanity.com/paper/2505.17771/) (2025)
- [A Point Cloud Generative Model via Tree-Structured Graph Convolutions for 3D Brain Shape Reconstruction](https://mlsanity.com/paper/2107.09923/) (2021)
- [Keypoint Cascade Voting for Point Cloud Based 6DoF Pose Estimation](https://mlsanity.com/paper/2210.08123/) (2022)
- [PointSCNet: Point Cloud Structure and Correlation Learning Based on Space Filling Curve-Guided Sampling](https://mlsanity.com/paper/2202.10251/) (2022)
- [SPNet: Multi-Shell Kernel Convolution for Point Cloud Semantic Segmentation](https://mlsanity.com/paper/2109.11610/) (2021)
- [MeT: A Graph Transformer for Semantic Segmentation of 3D Meshes](https://mlsanity.com/paper/2307.01115/) (2023)
- [Advancements in Point Cloud-Based 3D Defect Detection and Classification for Industrial Systems: A Comprehensive Survey](https://mlsanity.com/paper/2402.12923/) (2024)
- [Point-LN: A Lightweight Framework for Efficient Point Cloud Classification Using Non-Parametric Positional Encoding](https://mlsanity.com/paper/2501.14238/) (2025)
- [Freeview Sketching: View-Aware Fine-Grained Sketch-Based Image Retrieval](https://mlsanity.com/paper/2407.01810/) (2024)
- [CLIP-based Point Cloud Classification via Point Cloud to Image Translation](https://mlsanity.com/paper/2408.03545/) (2024)
- [Potential Convolution: Embedding Point Clouds into Potential Fields](https://mlsanity.com/paper/2104.01754/) (2021)
- [Deep and Shallow Covariance Feature Quantization for 3D Facial Expression Recognition](https://mlsanity.com/paper/2105.05708/) (2021)
- [Surface-biased Multi-Level Context 3D Object Detection](https://mlsanity.com/paper/2302.06291/) (2023)
- [Semantic Segmentation of Urban Textured Meshes Through Point Sampling](https://mlsanity.com/paper/2302.10635/) (2023)
- [Large-Language-Model Empowered Dose Volume Histogram Prediction for Intensity Modulated Radiotherapy](https://mlsanity.com/paper/2402.07167/) (2024)
- [Multi-view Structural Convolution Network for Domain-Invariant Point Cloud Recognition of Autonomous Vehicles](https://mlsanity.com/paper/2501.16289/) (2025)

## Abstract (arXiv)

Building discriminative representations for 3D data has been an important task in computer graphics and computer vision research. Convolutional Neural Networks (CNNs) have shown to operate on 2D images with great success for a variety of tasks. Lifting convolution operators to 3D (3DCNNs) seems like a plausible and promising next step. Unfortunately, the computational complexity of 3D CNNs grows cubically with respect to voxel resolution. Moreover, since most 3D geometry representations are boundary based, occupied regions do not increase proportionately with the size of the discretization, resulting in wasted computation. In this work, we represent 3D spaces as volumetric fields, and propose a novel design that employs field probing filters to efficiently extract features from them. Each field probing filter is a set of probing points --- sensors that perceive the space. Our learning algorithm optimizes not only the weights associated with the probing points, but also their locations, which deforms the shape of the probing filters and adaptively distributes them in 3D space. The optimized probing points sense the 3D space "intelligently", rather than operating blindly over the entire domain. We show that field probing is significantly more efficient than 3DCNNs, while providing state-of-the-art performance, on classification tasks for 3D object recognition benchmark datasets.