AI Agent Skills

Predict dense depth maps from sparse and noisy LiDAR frames guided by RGB images. (Ranked 1st place on KITTI) [MVA 2019]

Source code for the article "GroundGrid: LiDAR Point Cloud Ground Segmentation and Terrain Estimation"

LV-DOT: LiDAR-Visual Dynamic Obstacle Detection and Tracking (C++/Python/ROS)

DLL: Direct Lidar Localization

SSL_SLAM2: Lightweight 3-D Localization and Mapping for Solid-State LiDAR (mapping and localization separated) ICRA 2021

Real-time LIDAR-based Urban Road and Sidewalk detection for Autonomous Vehicles 🚗

VeloView performs real-time visualization and easy processing of live captured 3D LiDAR data from Velodyne sensors (Alpha Prime™, Puck™, Ultra Puck™, Puck Hi-Res™, Alpha…

Real-time people tracking using a 3D LIDAR

An unscented Kalman Filter implementation for fusing lidar and radar sensor measurements.

ROS 1 and ROS 2 node that converts sensor_msgs/PointCloud2 LIDAR data to nav_msgs/OccupancyGrid 2D map data based on intensity and / or height. 🤖

Implementation of SqueezeSegV2, Improved Model Structure and Unsupervised Domain Adaptation for Road-Object Segmentation from a LiDAR Point Cloud

[T-ASE 25] SegNet4D: Efficient Instance-Aware 4D Semantic Segmentation for LiDAR Point Cloud

[IROS 25] Leveraging Semantic Graphs for Efficient and Robust LiDAR SLAM

Visual Inertial Odometry (VIO) / Simultaneous Localization & Mapping (SLAM) using iSAM2 framework from the GTSAM library.

Simultaneous Visual Odometry, Object Detection, and Instance Segmentation

Make it Dense: Self-Supervised Geometric Scan Completion of Sparse 3D LiDAR Scans in Large Outdoor Environments