Point Pillars in a very famous 3D Object Detection Algorithm which got into light because of its fast inference speed on LiDAR generated point clouds. In this post, we will walk through its implementation code in TensorFlow.

If you want to get started on 3D Object Detection and more specifically on Point Pillars, I have a series of posts written on it just for that purpose. Also, going through the  Point Pillars paper directly will be really helpful to understand this post better, however, it’s not a prerequisite.

Github Repo for this tutorial: https://github.com/tyagi-iiitv/PointPillars. Thanks to Francesco Ferroni for the initial code.

Contents

• Defining Initial Configurations
• Building Point Pillars Network
• Loss functions
• Data Readers for Kitti Dataset
• Training
• Inference

#deep-learning #object-detection #machine-learning #data-science #tensorflow