Detection and Tracking
Object detection is one of the major lidar applications. The objects detected in lidar point cloud data are crucial for downstream workflows like tracking and labeling. Lidar Toolbox™ provides the object detection CNN PointPillars for developing custom object detection models.
Lidar Toolbox provides detection and tracking workflows for vehicles and road lanes. Most of the tracking workflows use the joint probabilistic data association (JPDA) tracker.
Geometric Shape Fitting
Load Training Data
|Ground truth label data|
|Combine data from multiple datastores|
|Datastore with custom file reader|
|Datastore for bounding box label data|
Augment and Preprocess Training Data
|Create randomized 3-D affine transformation|
|Apply geometric transformation to bounding boxes|
|Transform 3-D point cloud|
|PointPillars object detector|
|Train PointPillars object detector|
|Detect objects using PointPillars object detector|
Understand how to use point clouds for deep learning.
Define PointPillars network and learn how to perform object detection using the same.
Datastores for Deep Learning (Deep Learning Toolbox)
Learn how to use datastores in deep learning applications.
List of Deep Learning Layers (Deep Learning Toolbox)
Discover all the deep learning layers in MATLAB®.
Compare visualization functions.