# L2G Auto-encoder: Understanding Point Clouds by Local-to-Global Reconstruction with Hierarchical Self-Attention

arXiv: https://arxiv.org/abs/1908.00720 · HTML: https://mlsanity.com/paper/1908.00720/

## Key facts

- Cited by 32 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 2 Aug 2019.

## Built on by (most cited first, 30 of 32)

- [Attention Mechanisms in Computer Vision: A Survey](https://mlsanity.com/paper/2111.07624/) (2021)
- [SnowflakeNet: Point Cloud Completion by Snowflake Point Deconvolution with Skip-Transformer](https://mlsanity.com/paper/2108.04444/) (2021)
- [M3DeTR: Multi-representation, Multi-scale, Mutual-relation 3D Object Detection with Transformers](https://mlsanity.com/paper/2104.11896/) (2021)
- [PMP-Net++: Point Cloud Completion by Transformer-Enhanced Multi-step Point Moving Paths](https://mlsanity.com/paper/2202.09507/) (2022)
- [Unsupervised Point Cloud Representation Learning with Deep Neural Networks: A Survey](https://mlsanity.com/paper/2202.13589/) (2022)
- [Self-Contrastive Learning with Hard Negative Sampling for Self-supervised Point Cloud Learning](https://mlsanity.com/paper/2107.01886/) (2021)
- [Cycle4Completion: Unpaired Point Cloud Completion using Cycle Transformation with Missing Region Coding](https://mlsanity.com/paper/2103.07838/) (2021)
- [Snowflake Point Deconvolution for Point Cloud Completion and Generation with Skip-Transformer](https://mlsanity.com/paper/2202.09367/) (2022)
- [RandomRooms: Unsupervised Pre-training from Synthetic Shapes and Randomized Layouts for 3D Object Detection](https://mlsanity.com/paper/2108.07794/) (2021)
- [Learning a More Continuous Zero Level Set in Unsigned Distance Fields through Level Set Projection](https://mlsanity.com/paper/2308.11441/) (2023)
- [Orienting Point Clouds with Dipole Propagation](https://mlsanity.com/paper/2105.01604/) (2021)
- [Unsupervised Learning of Fine Structure Generation for 3D Point Clouds by 2D Projection Matching](https://mlsanity.com/paper/2108.03746/) (2021)
- [From Voxel to Point: IoU-guided 3D Object Detection for Point Cloud with Voxel-to-Point Decoder](https://mlsanity.com/paper/2108.03648/) (2021)
- [3D-OAE: Occlusion Auto-Encoders for Self-Supervised Learning on Point Clouds](https://mlsanity.com/paper/2203.14084/) (2022)
- [Learning a Structured Latent Space for Unsupervised Point Cloud Completion](https://mlsanity.com/paper/2203.15580/) (2022)
- [RI-MAE: Rotation-Invariant Masked AutoEncoders for Self-Supervised Point Cloud Representation Learning](https://mlsanity.com/paper/2409.00353/) (2024)
- [Parts2Words: Learning Joint Embedding of Point Clouds and Texts by Bidirectional Matching between Parts and Words](https://mlsanity.com/paper/2107.01872/) (2021)
- [KSS-ICP: Point Cloud Registration based on Kendall Shape Space](https://mlsanity.com/paper/2211.02807/) (2022)
- [CP-Net: Contour-Perturbed Reconstruction Network for Self-Supervised Point Cloud Learning](https://mlsanity.com/paper/2201.08215/) (2022)
- [Attention-based Transformation from Latent Features to Point Clouds](https://mlsanity.com/paper/2112.05324/) (2021)
- [Hierarchical View Predictor: Unsupervised 3D Global Feature Learning through Hierarchical Prediction among Unordered Views](https://mlsanity.com/paper/2108.03743/) (2021)
- [PointVST: Self-Supervised Pre-training for 3D Point Clouds via View-Specific Point-to-Image Translation](https://mlsanity.com/paper/2212.14197/) (2022)
- [Attention Models for Point Clouds in Deep Learning: A Survey](https://mlsanity.com/paper/2102.10788/) (2021)
- [Data Augmentation-free Unsupervised Learning for 3D Point Cloud Understanding](https://mlsanity.com/paper/2210.02798/) (2022)
- [D-Net: Learning for Distinctive Point Clouds by Self-Attentive Point Searching and Learnable Feature Fusion](https://mlsanity.com/paper/2305.05842/) (2023)
- [Extending Multi-Sense Word Embedding to Phrases and Sentences for Unsupervised Semantic Applications](https://mlsanity.com/paper/2103.15330/) (2021)
- [Synthesizing realistic sand assemblies with denoising diffusion in latent space](https://mlsanity.com/paper/2306.04411/) (2023)
- [Learning Local Pattern Modularization for Point Cloud Reconstruction from Unseen Classes](https://mlsanity.com/paper/2408.14279/) (2024)
- [Joint Data and Feature Augmentation for Self-Supervised Representation Learning on Point Clouds](https://mlsanity.com/paper/2211.01184/) (2022)
- [Masked Clustering Prediction for Unsupervised Point Cloud Pre-training](https://mlsanity.com/paper/2508.08910/) (2025)

## Abstract (arXiv)

Auto-encoder is an important architecture to understand point clouds in an encoding and decoding procedure of self reconstruction. Current auto-encoder mainly focuses on the learning of global structure by global shape reconstruction, while ignoring the learning of local structures. To resolve this issue, we propose Local-to-Global auto-encoder (L2G-AE) to simultaneously learn the local and global structure of point clouds by local to global reconstruction. Specifically, L2G-AE employs an encoder to encode the geometry information of multiple scales in a local region at the same time. In addition, we introduce a novel hierarchical self-attention mechanism to highlight the important points, scales and regions at different levels in the information aggregation of the encoder. Simultaneously, L2G-AE employs a recurrent neural network (RNN) as decoder to reconstruct a sequence of scales in a local region, based on which the global point cloud is incrementally reconstructed. Our outperforming results in shape classification, retrieval and upsampling show that L2G-AE can understand point clouds better than state-of-the-art methods.