This repo complements a ROS 2 course with hands-on C++/Python exercises, Gazebo simulation, and real robot control focusing on localization, mapping, and obstacle avoidance.
MASt3R-SLAM integrates a pretrained 3D reconstruction model as a geometry prior in a dense SLAM pipeline, enabling real-time tracking and mapping without classical bundle adjustment or depth sensors.
MonoGS rethinks monocular SLAM by replacing point-cloud maps with differentiable 3D Gaussian splatting, enabling real-time dense reconstruction and camera tracking in a unified pipeline.
OverlapNet uses Siamese networks on 2D range images from 3D LiDAR to detect loop closures by predicting overlap and relative yaw angle simultaneously. Practical demos included.
DROID-W builds on DROID-SLAM to handle dynamic scenes in-the-wild by jointly estimating camera pose, scene structure, and dynamic uncertainty using Lie group optimization and metric depth estimation.
Explore the ZED SDK, a C++ library for real-time stereo vision, SLAM, and spatial mapping with GPU acceleration and zero-copy CUDA interoperability for edge robotics.
MR.ScaleMaster fuses scale-ambiguous monocular SLAM trajectories from multiple robots using Sim(3) graph optimization, enabling heterogeneous SLAM frontends and consistent global maps.
PromptHMR adapts SAM’s promptable design to 3D human mesh recovery, integrating SLAM, pose detection, and SMPL models into a unified pipeline for monocular images and videos.