# FoldingNet: Point Cloud Auto-encoder via Deep Grid Deformation

arXiv: https://arxiv.org/abs/1712.07262 · HTML: https://mlsanity.com/paper/1712.07262/

## Key facts

- Cited by 24 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 19 Dec 2017.

## Built on by (most cited first, 24 of 24)

- [PoinTr: Diverse Point Cloud Completion with Geometry-Aware Transformers](https://mlsanity.com/paper/2108.08839/) (2021)
- [A Comprehensive Taxonomy for Explainable Artificial Intelligence: A Systematic Survey of Surveys on Methods and Concepts](https://mlsanity.com/paper/2105.07190/) (2021)
- [AdaPoinTr: Diverse Point Cloud Completion with Adaptive Geometry-Aware Transformers](https://mlsanity.com/paper/2301.04545/) (2023)
- [Point-Cloud Completion with Pretrained Text-to-image Diffusion Models](https://mlsanity.com/paper/2306.10533/) (2023)
- [Transformers in 3D Point Clouds: A Survey](https://mlsanity.com/paper/2205.07417/) (2022)
- [PvDeConv: Point-Voxel Deconvolution for Autoencoding CAD Construction in 3D](https://mlsanity.com/paper/2101.04493/) (2021)
- [3DRIMR: 3D Reconstruction and Imaging via mmWave Radar based on Deep Learning](https://mlsanity.com/paper/2108.02858/) (2021)
- [Neural Vector Fields: Implicit Representation by Explicit Learning](https://mlsanity.com/paper/2303.04341/) (2023)
- [Multi-class point cloud completion networks for 3D cardiac anatomy reconstruction from cine magnetic resonance images](https://mlsanity.com/paper/2307.08535/) (2023)
- [Explicitly Guided Information Interaction Network for Cross-modal Point Cloud Completion](https://mlsanity.com/paper/2407.02887/) (2024)
- [Modeling 3D cardiac contraction and relaxation with point cloud deformation networks](https://mlsanity.com/paper/2307.10927/) (2023)
- [Online Adaptation for Implicit Object Tracking and Shape Reconstruction in the Wild](https://mlsanity.com/paper/2111.12728/) (2021)
- [Free-form 3D Scene Inpainting with Dual-stream GAN](https://mlsanity.com/paper/2212.08464/) (2022)
- [Building-PCC: Building Point Cloud Completion Benchmarks](https://mlsanity.com/paper/2404.15644/) (2024)
- [Neural Rendering of Humans in Novel View and Pose from Monocular Video](https://mlsanity.com/paper/2204.01218/) (2022)
- [Multi-objective point cloud autoencoders for explainable myocardial infarction prediction](https://mlsanity.com/paper/2307.11017/) (2023)
- [Overlap Bias Matching is Necessary for Point Cloud Registration](https://mlsanity.com/paper/2308.09364/) (2023)
- [Neural Vector Fields: Generalizing Distance Vector Fields by Codebooks and Zero-Curl Regularization](https://mlsanity.com/paper/2309.01512/) (2023)
- [Flexible-weighted Chamfer Distance: Enhanced Objective Function for Point Cloud Completion](https://mlsanity.com/paper/2505.14218/) (2025)
- [Joint stereo 3D object detection and implicit surface reconstruction](https://mlsanity.com/paper/2111.12924/) (2021)
- [HeartFormer: Semantic-Aware Dual-Structure Transformers for 3D Four-Chamber Cardiac Point Cloud Reconstruction](https://mlsanity.com/paper/2512.00264/) (2025)
- [PPC-MT: Parallel Point Cloud Completion with Mamba-Transformer Hybrid Architecture](https://mlsanity.com/paper/2603.00870/) (2026)
- [Unlocking Hidden Potential in Point Cloud Networks with Attention-Guided Grouping-Feature Coordination](https://mlsanity.com/paper/2509.16639/) (2025)
- [DANCE: Density-agnostic and Class-aware Network for Point Cloud Completion](https://mlsanity.com/paper/2511.07978/) (2025)

## Abstract (arXiv)

Recent deep networks that directly handle points in a point set, e.g., PointNet, have been state-of-the-art for supervised learning tasks on point clouds such as classification and segmentation. In this work, a novel end-to-end deep auto-encoder is proposed to address unsupervised learning challenges on point clouds. On the encoder side, a graph-based enhancement is enforced to promote local structures on top of PointNet. Then, a novel folding-based decoder deforms a canonical 2D grid onto the underlying 3D object surface of a point cloud, achieving low reconstruction errors even for objects with delicate structures. The proposed decoder only uses about 7% parameters of a decoder with fully-connected neural networks, yet leads to a more discriminative representation that achieves higher linear SVM classification accuracy than the benchmark. In addition, the proposed decoder structure is shown, in theory, to be a generic architecture that is able to reconstruct an arbitrary point cloud from a 2D grid. Our code is available at http://www.merl.com/research/license#FoldingNet