SFM-AR-Visual-SLAM¶
Visual SLAM¶
GSLAM¶
General SLAM Framework which supports feature based or direct method and different sensors including monocular camera, RGB-D sensors or any other input types can be handled. https://github.com/zdzhaoyong/GSLAM
OKVIS: Open Keyframe-based Visual-Inertial SLAM¶
Uncertainty-aware Receding Horizon Exploration and Mapping Planner¶
S-PTAM: Stereo Parallel Tracking and Mapping¶
mcptam¶
MCPTAM is a set of ROS nodes for running Real-time 3D Visual Simultaneous Localization and Mapping (SLAM) using Multi-Camera Clusters. It includes tools for calibrating both the intrinsic and extrinsic parameters of the individual cameras within the rigid camera rig.
FAB-MAP¶
visual place recognition algorithm https://github.com/arrenglover/openfabmap
maplab¶
An Open Framework for Research in Visual-inertial Mapping and Localization https://github.com/ethz-asl/maplab from Roland Siegwart
OpenVSLAM: Versatile Visual SLAM Framework¶
SLAM with Apriltag¶
https://github.com/berndpfrommer/tagslam ROS ready, bag file available
Fast Odometry and Scene Flow from RGB-D Cameras¶
https://github.com/MarianoJT88/Joint-VO-SF published in ICRA 2017
Real-Time Appearance-Based Mapping¶
http://wiki.ros.org/rtabmap_ros … Many Demos are available in the website with Several ROS bags
general and scalable framework for visual SLAM¶
https://github.com/strasdat/ScaViSLAM/
https://github.com/felixendres/rgbdslam_v2 ROS ready, It accompany a PHD thesis from TUM
SLAM in unstructed environments¶
Dense Visual Odometry and SLAM (dvo_slam)¶
Coslam: Collaborative visual slam in dynamic environments¶
Real-time dense visual SLAM system : ElasticFusion¶
https://github.com/mp3guy/ElasticFusion … it has nice gui and dataset , paper and video too .
Real-time dense visual SLAM¶
Deferred Triangulation SLAM¶
Based on PTAM and SLAM track 3d traingulated and 2d non triangulated features . https://github.com/plumonito/dtslam
Dense RGBD slam¶
M2SLAM: Visual SLAM with Memory Management for large-scale Environments¶
SceneLib2 - MonoSLAM open-source library¶
from oxford university c++ SLAM
https://github.com/hanmekim/SceneLib2
next best view planner¶
Dynamic RGB-D Encoder SLAM for a Differential-Drive Robot¶
https://github.com/ydsf16/dre_slam ROS kinetic, openCV 4.0, yolo v3, Ceres
DynaSLAM: Tracking, Mapping and Inpainting in Dynamic Scenes¶
https://github.com/BertaBescos/DynaSLAM
Augmented Reality¶
PTAM (Parallel Tracking and Mapping) :¶
ORB-SLAM: A Versatile and Accurate Monocular SLAM System¶
https://github.com/raulmur/ORB_SLAM ….
its modification : ORB-SLAM2 is a real-time SLAM library for Monocular, Stereo and RGB-D cameras https://github.com/raulmur/ORB_SLAM2
its modification to work on IOS : https://github.com/Thunderbolt-sx/ORB_SLAM_iOS
ORB-SLAM3 An Accurate Open-Source Library for Visual, Visual-Inertial and Multi-Map SLAM¶
REMODE (REgularized MOnocular Depth Estimation)¶
https://github.com/uzh-rpg/rpg_open_remode … Probabilistic, Monocular Dense Reconstruction in Real Time
Fast Semi-Direct Monocular Visual Odometry¶
Fast Semi-Direct Visual Odometry for Monocular, Wide Angle, and Multi-camera Systems¶
no loop closure or bundle adjustment http://rpg.ifi.uzh.ch/svo2.html
LSD-SLAM: Large-Scale Direct Monocular SLAM¶
https://github.com/tum-vision/lsd_slam
modification over the original package to work with rolling chatter camera ( cheap webcams) https://github.com/FirefoxMetzger/lsd_slam The change is mentioned in this video : https://www.youtube.com/watch?v=TZRICW6R24o
ROS wrapper for visolib¶
https://github.com/srv/viso2 It is supported till ROS-indigo.
Visual-Inertia-fusion-based Monocular dEnse mAppiNg¶
https://github.com/HKUST-Aerial-Robotics/VI-MEAN with paper and video ICRA 2017 , rosbag as well.
monocular object pose SLAM¶
DeepFactors: Real-Time Probabilistic Dense Monocular SLAM¶
LIDAR based¶
LIMO: Lidar-Monocular Visual Odometry¶
https://github.com/johannes-graeter/limo Virtual machine with all the dependencies is ready.
LiDAR-based real-time 3D localization and mapping¶
segmatch¶
https://github.com/ethz-asl/segmatch A 3D segment based loop-closure algorithm | ROS ready
LIO-SAM¶
https://github.com/TixiaoShan/LIO-SAM real-time lidar-inertial odometry
Visual Odometry¶
Dense Sparse odometry¶
monocular odometry algorithm¶
https://github.com/alejocb/dpptam Dense Piecewise Planar Tracking and Mapping from a Monocular Sequence IROS 2015
Stereo Visual odometry¶
https://github.com/rubengooj/StVO-PL Stereo Visual Odometry by combining point and line segment features
Monocular Motion Estimation on Manifolds¶
Visual Odometry Revisited: What Should Be Learnt?¶
paper + pytorch code: https://github.com/Huangying-Zhan/DF-VO
SimVODIS Simultaneous Visual Odometry, Object Detection, and Instance Segmentation¶
https://github.com/Uehwan/SimVODIS
Visual Inertial odometry¶
Kalibr¶
IMU camera calibration toolbox and more. https://github.com/ethz-asl/kalibr
Camera-to-IMU calibration toolbox https://github.com/hovren/crisp
ROVIO¶
Robust Visual Inertial Odometry https://github.com/ethz-asl/rovio
Robust Stereo Visual Inertial Odometry for Fast Autonomous Flight¶
A Robust and Versatile Monocular Visual-Inertial State Estimator¶
VINS modification for omnidirectional + Streo camera¶
Realtime Edge Based Inertial Visual Odometry for a Monocular Camera¶
https://github.com/JuanTarrio/rebvo Specially targetted to embedded hardware.
robocentric visual-inertial odometry¶
https://github.com/rpng/R-VIO Monocular camera + 6 DOF IMU
SFM¶
Structure from Motion (SfM) for Unordered Image Collections¶
Five Point , 6,7,8 algorithms¶
open geometrical vision https://github.com/marknabil/opengv
openSFM¶
Structure from Motion library written in Python on top of OpenCV. It has dockerfile for all installation on ubuntu 14.04 https://github.com/mapillary/OpenSfM
Unsupervised Learning of Depth and Ego-Motion from Video¶
An unsupervised learning framework for depth and ego-motion estimation from monocular videos https://github.com/tinghuiz/SfMLearner
CVPR 2015 Tutorial for open source SFM¶
Source material for the CVPR 2015 Tutorial: Open Source Structure-from-Motion https://github.com/mleotta/cvpr2015-opensfm
Unsupervised Learning of Depth and Ego-Motion from Video¶
concepts in matlab¶
http://vis.uky.edu/~stewe/FIVEPOINT/
SFMedu: A Matlab-based Structure-from-Motion System for Education https://github.com/jianxiongxiao/SFMedu
Lorenzo Torresani’s Structure from Motion Matlab code https://github.com/scivision/em-sfm
https://github.com/vrabaud/sfm_toolbox
OpenMVG C++ library https://github.com/openMVG/openMVG
collection of computer vision methods for solving geometric vision problems https://github.com/laurentkneip/opengv
Multiview Geometry Library in C++11¶
Quaternion Based Camera Pose Estimation From Matched Feature Points¶
https://sites.google.com/view/kavehfathian/code its paper : https://arxiv.org/pdf/1704.02672.pdf
Others :¶
SLAM with IMU on Android¶
IOS iphone 7 plus¶
Matlab¶
with some good documentation to how to read the image and so on from the kinect . https://github.com/AutoSLAM/SLAM
Datasets and benchmarking¶
Curated List of datasets:¶
https://github.com/youngguncho/awesome-slam-datasets
EuRoC MAV Dataset¶
http://projects.asl.ethz.ch/datasets/doku.php?id=kmavvisualinertialdatasets
visual-inertial datasets collected on-board a Micro Aerial Vehicle (MAV). The datasets contain stereo images, synchronized IMU measurements, and accurate motion and structure ground-truth.
TUM VI Benchmark for Evaluating Visual-Inertial Odometry¶
https://vision.in.tum.de/data/datasets/visual-inertial-dataset different scenes for evaluating VI odometry
Authentic Dataset for Visual-Inertial Odometry¶
challenging Visual Inertial Odometry benchmark¶
https://daniilidis-group.github.io/penncosyvio/ from Pennsylvania, published in ICRA2017
ICL NIUM¶
https://www.doc.ic.ac.uk/~ahanda/VaFRIC/iclnuim.html benchmarking RGB-D, Visual Odometry and SLAM algorithms
Benchmarking Pose Estimation Algorithms¶
https://sites.google.com/view/kavehfathian/code/benchmarking-pose-estimation-algorithms
Toolbox for quantitative trajectory evaluation of VO/VIO¶
Photorealistic Simulator for VIO testing/benchmarking¶
Machine Learning/ Deep learning based¶
Learning monocular visual odometry with dense 3D mapping from dense 3D flow
DeepVO: A Deep Learning approach for Monocular Visual Odometry
Survey papers and articles¶
Survey with year,sensor used and best practice
Imperial college ICCV 2015 workshop
Deep Auxiliary Learning for Visual Localization and Odometry
follow :¶
Robotics and Perception Group¶
TUM VISION¶
Another Curated list¶
for SFM, 3D reconstruction and V-SLAM https://github.com/openMVG/awesome_3DReconstruction_list