Two Heads are Better than One: Geometric-Latent Attention for Point Cloud Classification and Segmentation
Not on the map. Why?
Abstract (arXiv)
We present an innovative two-headed attention layer that combines geometric and latent features to segment a 3D scene into semantically meaningful subsets. Each head combines local and global information, using either the geometric or latent features, of a neighborhood of points and uses this information to learn better local relationships. This Geometric-Latent attention layer (Ge-Latto) is combined with a sub-sampling strategy to capture global features. Our method is invariant to permutation thanks to the use of shared-MLP layers, and it can also be used with point clouds with varying densities because the local attention layer does not depend on the neighbor order. Our proposal is simple yet robust, which allows it to achieve competitive results in the ShapeNetPart and ModelNet40 datasets, and the state-of-the-art when segmenting the complex dataset S3DIS, with 69.2% IoU on Area 5, and 89.7% overall accuracy using K-fold cross-validation on the 6 areas.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 34 of its 40 references that are in the map. First posted on arXiv 30 Oct 2021.
Builds on
34 of 40 references.
References · 34
- 2017Attention Is All You Needcited by 59,932
- 2015Deep Residual Learning for Image Recognitioncited by 51,809
- 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
- 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
- 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
- 2016Pyramid Scene Parsing Networkcited by 2,472
- 2018Relational inductive biases, deep learning, and graph networkscited by 1,341
- 2017VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detectioncited by 1,323
- 2019KPConv: Flexible and Deformable Convolution for Point Cloudscited by 1,202
- 20194D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networkscited by 1,201
- 2018Self-Attention with Relative Position Representationscited by 1,112
- 2018PointCNN: Convolution On X-Transformed Pointscited by 764
22 more
- 2017Focal Loss for Dense Object Detectioncited by 749
- 2020PCT: Point cloud transformercited by 730
- 2018PointConv: Deep Convolutional Networks on 3D Point Cloudscited by 656
- 2019RandLA-Net: Efficient Semantic Segmentation of Large-Scale Point Cloudscited by 584
- 2020Searching Efficient 3D Architectures with Sparse Point-Voxel Convolutioncited by 422
- 2017Large-scale Point Cloud Semantic Segmentation with Superpoint Graphscited by 384
- 2016OctNet: Learning Deep 3D Representations at High Resolutionscited by 340
- 2020Exploring Self-attention for Image Recognitioncited by 313
- 2018SO-Net: Self-Organizing Network for Point Cloud Analysiscited by 210
- 2017SEGCloud: Semantic Segmentation of 3D Point Cloudscited by 178
- 2018Tangent Convolutions for Dense Prediction in 3Dcited by 173
- 2021Walk in the Cloud: Learning Curves for Point Clouds Shape Analysiscited by 172
- 2017Escape from Cells: Deep Kd-Networks for the Recognition of 3D Point Cloud Modelscited by 142
- 2020A Closer Look at Local Aggregation Operators in Point Cloud Analysiscited by 87
- 2021Semantic Segmentation for Real Point Cloud Scenes via Bilateral Augmentation and Adaptive Fusioncited by 71
- 2018VV-Net: Voxel VAE Net with Group Convolutions for Point Cloud Segmentationcited by 64
- 2020FPConv: Learning Local Flattening for Point Convolutioncited by 55
- 2017O-CNN: Octree-based Convolutional Neural Networks for 3D Shape Analysiscited by 47
- 2018TextureNet: Consistent Local Parametrizations for Learning from High-Resolution Signals on Meshescited by 45
- 2018H-CNN: Spatial Hashing Based CNN for 3D Shape Analysiscited by 10
- 2018TrimBot2020: an outdoor robot for automatic gardeningcited by 9
- 2018Hybrid Multi-camera Visual Servoing to Moving Targetcited by 2
Built on by
Citing papers per year: 2024: 1.
1 paper.