Posts

A novel two-direction vision transformer network to fuse camera and LiDAR data for semantic segmentation in autonomous driving.

This project aims to develop a FCN-based network for semantic segmentation in autonomous driving on the iseAuto dataset.

Autoware is one of the most popular open-source project that provides a complete set of solutions for autonomous driving, including mapping, localization, prediction, planning and controlling. This post introduces the detailed procedures to implement the basic Autoware self-driving modules in Boxbot robot.

An open-source training dataset for perception of autonomous vehicles. Collected by iseAuto shuttle (first L4 self-driving shuttle in Estonia). Featured with the challenging wet and dark scenarios

This post describes the detailed steps of how we calibrate LiDAR and camera for iseAuto shuttle. The main purpose is getting two sensors’ extrinsic matrix, which is essential for any further LiDAR-camera fusion applications.