Build and deploy AI systems that derive information from point clouds. Train, test, and deploy deep learning networks on point clouds for object detection and semantic segmentation.
With MATLAB and Simulink, you can:
- Preprocess point clouds for applying deep learning algorithms
- Use the Lidar Labeler app to label point clouds for object detection
- Handle large amounts of data for training, testing, and validation with datastores
- Generate C/C++ and CUDA codes for deep learning workflows for semantic segmentation and object detection on point cloud data
Why Use Deep Learning on Point Clouds?
Point Cloud Semantic Segmentation
Apply deep learning algorithms to segment point clouds. Train, test, and evaluate semantic segmentation networks, including RandLA-Net, Segment Anything Model (SAM), PointSeg, SqueezeSegV2, and PointNet++ on point cloud data.
Object Detection on Point Clouds
Detect and fit oriented bounding boxes around objects in point clouds and use them for object tracking or lidar labeling workflows. Design, train, and evaluate robust detectors such as PointPillars and Voxel R-CNN networks.
Point Cloud Labeling using Multi Sensor Labeler
Label point clouds for training deep learning models. Apply built-in or custom algorithms to automate point cloud labeling with the Multi Sensor Labeler app and evaluate automation algorithm performance.
Deployment
Generate CUDA® MEX code for networks like PointPillars, SqueezeSegV2, and PointNet++ to deploy point cloud segmentation or object detection algorithms on GPUs.
Using Deep Learning for Point Cloud with MATLAB
Products
Learn about the products used with deep learning for point cloud applications.
