MichaelBeechan/VO-SLAM-Review MichaelBeechan/VO-SLAM-Review

  • Stars
    star
    219
  • Rank 181,133 (Top 4 %)
  • Language
  • License
    Apache License 2.0
  • Created over 5 years ago
  • Updated about 2 years ago
Twitter logo Share LinkedIn logo Share Facebook logo Share
MichaelBeechan/VO-SLAM-Review
github

MichaelBeechan

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

SLAM is mainly divided into two parts: the front end and the back end. The front end is the visual odometer (VO), which roughly estimates the motion of the camera based on the information of adjacent images and provides a good initial value for the back end.The implementation methods of VO can be divided into two categories according to whether features are extracted or not: feature point-based methods, and direct methods without feature points. VO based on feature points is stable and insensitive to illumination and dynamic objects

Visual Odometry（VO）-SLAM-Review

SLAM is mainly divided into two parts: the front end and the back end. The front end is the visual odometer (VO), which roughly estimates the motion of the camera based on the information of adjacent images and provides a good initial value for the back end.The implementation methods of VO can be divided into two categories according to whether features are extracted or not: feature point-based methods, and direct methods without feature points. VO based on feature points is stable and insensitive to illumination and dynamic objects

=======================================================================

github：https://github.com/MichaelBeechan

CSDN：https://blog.csdn.net/u011344545

=======================================================================

SALM review paper download：

https://download.csdn.net/download/u011344545/10850261

1、Visual Odometry or VSLAM

2、Visual Inertial Odometry or VIO-SLAM

3、Based CNN（Net VO or Net VSLAM）

4、Lidar Visual odometry or Lidar SLAM

5、Semanitc SLAM

6、Datasets

7、Libraries

OF-VO：Robust and Efficient Stereo Visual Odometry Using Points and Feature Optical Flow

Code：https://github.com/MichaelBeechan/MyStereoLibviso2

SLAMBook

Paper：14 Lectures on Visual SLAM: From Theory to Practice,

Code：https://github.com/gaoxiang12/slambook

SLAMBook2

Code：https://github.com/gaoxiang12/slambook2

SVO: Fast Semi-Direct Monocular Visual Odometry

Paper：http://rpg.ifi.uzh.ch/docs/ICRA14_Forster.pdf

Video: http://youtu.be/2YnIMfw6bJY

Code：https://github.com/uzh-rpg/rpg_svo

Robust Odometry Estimation for RGB-D Cameras

Real-Time Visual Odometry from Dense RGB-D Images

Paper：http://www.cs.nuim.ie/research/vision/data/icra2013/Whelan13icra.pdf

Code：https://github.com/tum-vision/dvo

Parallel Tracking and Mapping for Small AR Workspaces

Paper：https://cse.sc.edu/~yiannisr/774/2015/ptam.pdf

http://www.robots.ox.ac.uk/ActiveVision/Papers/klein_murray_ismar2007/klein_murray_ismar2007.pdf

Code：https://github.com/Oxford-PTAM/PTAM-GPL

ORBSLAM

Code3：https://github.com/MichaelBeechan/ORBSLAM3

Paper3：ORB-SLAM3: An Accurate Open-Source Library for Visual, Visual-Inertial and Multi-Map SLAM：https://arxiv.org/pdf/2007.11898.pdf

Code2：https://github.com/raulmur/ORB_SLAM2

Paper2：ORB-SLAM2: an Open-Source SLAM System for Monocular, Stereo and RGB-D Cameras：https://arxiv.org/pdf/1610.06475.pdf

Code1：https://github.com/raulmur/ORB_SLAM

Paper1:ORB-SLAM: A Versatile and Accurate Monocular SLAM System:https://arxiv.org/pdf/1502.00956.pdf

A ROS Implementation of the Mono-Slam Algorithm

Paper：https://www.researchgate.net/publication/269200654_A_ROS_Implementation_of_the_Mono-Slam_Algorithm

Code：https://github.com/rrg-polito/mono-slam

DTAM: Dense tracking and mapping in real-time

Paper：https://ieeexplore.ieee.org/document/6126513

Code：https://github.com/anuranbaka/OpenDTAM

LSD-SLAM: Large-Scale Direct Monocular SLAM

Paper：http://pdfs.semanticscholar.org/c13c/b6dfd26a1b545d50d05b52c99eb87b1c82b2.pdf

https://vision.in.tum.de/research/vslam/lsdslam

Code：https://github.com/tum-vision/lsd_slam

RGBD-Odometry (Visual Odometry based RGB-D images)

Real-Time Visual Odometry from Dense RGB-D Images

Code：https://github.com/tzutalin/OpenCV-RgbdOdometry

Paper：http://www.computer.org/csdl/proceedings/iccvw/2011/0063/00/06130321.pdf

Py-MVO: Monocular Visual Odometry using Python

Code：https://github.com/Transportation-Inspection/visual_odometry

Video：https://www.youtube.com/watch?v=E8JK19TmTL4&feature=youtu.be

Stereo-Odometry-SOFT

MATLAB Implementation of Visual Odometry using SOFT algorithm

Code：https://github.com/Mayankm96/Stereo-Odometry-SOFT

Paper：https://ieeexplore.ieee.org/document/7324219

GF_ORB_SLAM：Good Feature Matching: Towards Accurate, Robust VO/VSLAM with Low Latency

Paper：https://arxiv.org/pdf/2001.00714.pdf

Code：https://github.com/ivalab/GF_ORB_SLAM

monoVO-python

Code1：https://github.com/uoip/monoVO-pythone：https://github.com/uoip/monoVO-python

Code2：https://github.com/yueying/LearningVO

DVO：Robust Odometry Estimation for RGB-D Cameras

Code：https://github.com/tum-vision/dvo

https://vision.in.tum.de/data/software/dvo

Paper：https://www.researchgate.net/publication/221430091_Real-time_visual_odometry_from_dense_RGB-D_images

Dense Visual Odometry and SLAM (dvo_slam)

Code：https://github.com/tum-vision/dvo_slam

https://vision.in.tum.de/data/software/dvo

Paper：https://www.researchgate.net/publication/261353146_Dense_visual_SLAM_for_RGB-D_cameras

REVO：Robust Edge-based Visual Odometry

Combining Edge Images and Depth Maps for Robust Visual Odometry

Robust Edge-based Visual Odometry using Machine-Learned Edges

Code：https://github.com/fabianschenk/REVO

Paper：https://graz.pure.elsevier.com/

xivo

X Inertial-aided Visual Odometry

Code：https://github.com/ucla-vision/xivo

Paper：XIVO: X Inertial-aided Visual Odometry and Sparse Mapping

PaoPaoRobot

Code：https://github.com/PaoPaoRobot

ygz-slam

Code：https://github.com/PaoPaoRobot/ygz-slam

https://github.com/gaoxiang12/ygz-stereo-inertial

https://github.com/gaoxiang12/ORB-YGZ-SLAM

https://www.ctolib.com/generalized-intelligence-GAAS.html#5-ygz-slam

RTAB MAP

Online Global Loop Closure Detection for Large-Scale Multi-Session Graph-Based SLAM, 2014 Appearance-Based Loop Closure Detection for Online Large-Scale and Long-Term Operation, 2013

MYNT-EYE

Code：https://github.com/slightech

Kintinuous

Real-time Large Scale Dense RGB-D SLAM with Volumetric Fusion

Deformation-based Loop Closure for Large Scale Dense RGB-D SLAM

Robust Real-Time Visual Odometry for Dense RGB-D Mapping

Kintinuous: Spatially Extended KinectFusion

A method and system for mapping an environment

Code：https://github.com/mp3guy/Kintinuous

ElasticFusion

ElasticFusion: Dense SLAM Without A Pose Graph

ElasticFusion: Real-Time Dense SLAM and Light Source Estimation

Paper：http://www.thomaswhelan.ie/Whelan16ijrr.pdf http://thomaswhelan.ie/Whelan15rss.pdf

Code：https://github.com/mp3guy/ElasticFusion

Co-Fusion：Real-time Segmentation, Tracking and Fusion of Multiple Objects

Paper：http://visual.cs.ucl.ac.uk/pubs/cofusion/index.html

R-VIO：Robocentric Visual-Inertial Odometry

(Kimera-VIO is a Visual Inertial Odometry pipeline for accurate State Estimation from Stereo + IMU data.)

Code：https://github.com/rpng/R-VIO

Paper：https://arxiv.org/abs/1805.04031

Kimera-VIO: Open-Source Visual Inertial Odometry

Code：https://github.com/MIT-SPARK/Kimera-VIO

Paper：https://arxiv.org/abs/1910.02490

Kimera: an Open-Source Library for Real-Time Metric-Semantic Localization and Mapping

ADVIO: An Authentic Dataset for Visual-Inertial Odometry

Code：https://github.com/AaltoVision/ADVIO

Paper：https://arxiv.org/abs/1807.09828

Data：https://zenodo.org/record/1476931#.XgCvYVIza00

MSCKF_VIO：Robust Stereo Visual Inertial Odometry for Fast Autonomous Flight

Paper：https://arxiv.org/abs/1712.00036

Code：https://github.com/KumarRobotics/msckf_vio

Kimera-VIO: Open-Source Visual Inertial Odometry

Kimera: an Open-Source Library for Real-Time Metric-Semantic Localization and Mapping

Code：https://github.com/MIT-SPARK/Kimera-VIO

Paper：https://arxiv.org/abs/1910.02490

LIBVISO2: C++ Library for Visual Odometry 2

Paper：http://www.cvlibs.net/software/libviso/

Code：https://github.com/srv/viso2

Stereo Visual SLAM for Mobile Robots Navigation

A constant-time SLAM back-end in the continuum between global mapping and submapping: application to visual stereo SLAM

Paper：http://mapir.uma.es/famoreno/papers/thesis/FAMD_thesis.pdf

Code：https://github.com/famoreno/stereo-vo

Combining Edge Images and Depth Maps for Robust Visual Odometry

Robust Edge-based Visual Odometry using Machine-Learned Edges(REVO)

Paper：https://graz.pure.elsevier.com/

Code：https://github.com/fabianschenk/REVO

HKUST Aerial Robotics Group

VINS-Mono: A Robust and Versatile Monocular Visual-Inertial State Estimator

Paper：https://arxiv.org/pdf/1708.03852.pdf

Code：https://github.com/HKUST-Aerial-Robotics/VINS-Mono

VINS-Fusion:Online Temporal Calibration for Monocular Visual-Inertial Systems

Paper：https://arxiv.org/pdf/1808.00692.pdf

Code：https://github.com/HKUST-Aerial-Robotics/VINS-Fusion

Monocular Visual-Inertial State Estimation for Mobile Augmented Reality

Paper：https://ieeexplore.ieee.org/document/8115400

Code：https://github.com/HKUST-Aerial-Robotics/VINS-Mobile

Computer Vision Group TUM Department of Informatics Technical University of Munich

DSO: Direct Sparse Odometry

Code：https://github.com/JingeTu/StereoDSO

Visual-Inertial DSOhttps://vision.in.tum.de/research/vslam/vi-dso

DVSO：https://vision.in.tum.de/research/vslam/dvso

DSO with Loop-closure and Sim(3) pose graph optimization：https://vision.in.tum.de/research/vslam/ldso

Stereo odometry based on careful feature selection and tracking

Paper：https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7324219

Code：https://github.com/Mayankm96/Stereo-Odometry-SOFT

OKVIS: Open Keyframe-based Visual-Inertial SLAM

Code：https://github.com/gaoxiang12/okvis

Trifo-VIO: Robust and Efficient Stereo Visual Inertial Odometry using Points and Lines

Paper：https://arxiv.org/pdf/1803.02403.pdf

Code：https://github.com/UMiNS/Trifocal-tensor-VIO

PL-VIO: Tightly-Coupled Monocular Visual–Inertial Odometry Using Point and Line Features

Paper：https://www.mdpi.com/1424-8220/18/4/1159/html

Overview of visual inertial navigation

A Review of Visual-Inertial Simultaneous Localization and Mapping from Filtering-Based and Optimization-Based Perspectives：

https://ieeexplore.ieee.org/document/5423178

https://www.mdpi.com/2218-6581/7/3/45

VINet: Visual-Inertial Odometry as a Sequence-to-Sequence Learning Problem

Paper：https://arxiv.org/abs/1701.08376

Code：https://github.com/HTLife/VINet

DeepVO: Towards End-to-End Visual Odometry with Deep Recurrent Convolutional Neural Networks

Code：https://github.com/ildoonet/deepvo

https://github.com/sladebot/deepvo

https://github.com/themightyoarfish/deepVO

https://github.com/fshamshirdar/DeepVO (pytorch)

Paper：http://www.cs.ox.ac.uk/files/9026/DeepVO.pdf

UnDeepVO: Implementation of Monocular Visual Odometry through Unsupervised Deep Learning

Code：https://github.com/drmaj/UnDeepVO

Paper：https://arxiv.org/pdf/1709.06841.pdf

SfM-Net: SfM-Net: Learning of Structure and Motion from Video

Code： https://github.com/waxz/sfm_net

Paper： https://arxiv.org/pdf/1704.07804v1.pdf

CNN-SLAM: CNN-SLAM: Real-time dense monocular SLAM with learned depth prediction

Code： https://github.com/iitmcvg/CNN_SLAM

Paper：https://arxiv.org/pdf/1704.03489.pdf

PoseNet: Posenet: A convolutional network for real-time 6-dof camera relocalization(ICCV2015)

Code：https://github.com/alexgkendall/caffe-posenet or https://github.com/kentsommer/tensorflow-posenet

Paper：https://arxiv.org/pdf/1505.07427.pdf or https://arxiv.org/pdf/1509.05909.pdf

VidLoc: VidLoc: 6-doF video-clip relocalization

Code: https://github.com/futurely/deep-camera-relocalization

Paper: https://arxiv.org/pdf/1702.06521.pdf

NetVLAD: NetVLAD: CNN architecture for weakly supervised place recognition(CVPR2016)

Code: https://github.com/Relja/netvlad (Matlab) or https://github.com/lyakaap/NetVLAD-pytorch

Paper: https://arxiv.org/pdf/1511.07247.pdf

DeMoN: Depth and Motion Network for Learning Monocular Stereo（CVPR2017）

Code: https://github.com/lmb-freiburg/demon

Paper: https://arxiv.org/pdf/1612.02401v2.pdf

Learned Stereo Machine

Code: https://github.com/akar43/lsm

Paper: https://arxiv.org/pdf/1708.05375.pdf

SfMLearner: Unsupervised Learning of Depth and Ego-Motion from Video

Code: https://github.com/tinghuiz/SfMLearner

Paper: https://arxiv.org/pdf/1704.07813.pdf

Toward Geometric Deep SLAM

Code: UNopen(https://github.com/mtrasobaresb)

Paper: https://arxiv.org/pdf/1707.07410v1.pdf

Neural SLAM : Learning to Explore with External Memory

Code: UNopen

Paper: https://arxiv.org/pdf/1706.09520.pdf

PoseConvGRU: A Monocular Approach for Visual Ego-motion Estimation by Learning(2019)

Code: UNopen

Paper: https://arxiv.org/pdf/1906.08095.pdf

Semi-Dense 3D Semantic Mapping from Monocular SLAM(2016)

Code: UNopen

Paper: https://arxiv.org/pdf/1611.04144.pdf

Image-based 3D Object Reconstruction: State-of-the-Art and Trends in the Deep Learning Era(2019)

Paper: https://arxiv.org/pdf/1906.06543.pdf

DeepMVS: DeepMVS: Learning Multi-view Stereopsis(CVPR2018)

Code: https://github.com/phuang17/DeepMVS

Paper: https://phuang17.github.io/DeepMVS/index.html

Paper: https://arxiv.org/pdf/1804.00650.pdf

MVSNet: Mvsnet: Depth inference for unstructured multi-view stereo(ECCV2018)

Code1: https://github.com/YoYo000/MVSNet

Code2: https://github.com/YoYo000/BlendedMVS

Paper: https://arxiv.org/pdf/1804.02505.pdf

PointMVSNet:Point-based Multi-view Stereo Network

Code: https://github.com/callmeray/PointMVSNet

Paper: https://arxiv.org/pdf/1908.04422.pdf

Recurrent MVSNet: Recurrent MVSNet for High-resolution Multi-view Stereo Depth Inference(CVPR2019)

Code: https://github.com/YoYo000/MVSNet

Paper: https://arxiv.org/pdf/1902.10556.pdf

(ESP-VO) End-to-End, Sequence-to-Sequence Probabilistic Visual Odometry through Deep Neural Networks

Code: https://github.com/espnet/espnet

https://www.seas.upenn.edu/~meam620/slides/kinematicsI.pdf

Lidar Visual odometry

Lidar-Monocular Visual Odometry

Code：https://github.com/johannes-graeter/limo

Paper：https://arxiv.org/pdf/1807.07524.pdf

RGBD and LIDAR

CAE-LO: LiDAR Odometry Leveraging Fully Unsupervised Convolutional Auto-Encoder for Interest Point Detection and Feature Description

Paper：https://arxiv.org/ftp/arxiv/papers/2001/2001.01354.pdf

Code：https://github.com/SRainGit/CAE-LO

Other open source projects

DynaSLAM A SLAM system robust in dynamic environments for monocular, stereo and RGB-D setups

openvslam A Versatile Visual SLAM Framework

cartographer

Code：https://github.com/googlecartographer/cartographer

Paper：https://google-cartographer.readthedocs.io/en/latest/

A-LOAM（Advanced implementation of LOAM）

LOAM: Lidar Odometry and Mapping in Real-time

Code1：https://github.com/HKUST-Aerial-Robotics/A-LOAM

Code2：https://github.com/cuitaixiang/LOAM_NOTED

Paper：http://roboticsproceedings.org/rss10/p07.pdf

SemanticFusion: Dense 3D semantic mapping with convolutional neural networks

Code: https://github.com/seaun163/semanticfusion

Paper: https://arxiv.org/pdf/1609.05130v2.pdf

ORB_SLAM2_SSD_Semantic

Code：https://github.com/Ewenwan/ORB_SLAM2_SSD_Semantic

Visual Semantic SLAM with Landmarks for Large-Scale Outdoor Environment

Paper：https://arxiv.org/pdf/2001.01028.pdf

Code：https://github.com/1989Ryan/Semantic_SLAM/

Datasets

Libraries

Basic vision and trasformation libraries

Thread-safe queue libraries

Loop detection

Graph Optimization

Map library

Tools

More Repositories

1

ThunderNet-Review

Real-time generic object detection on mobile platforms is a crucial but challenging computer vision task. However, previous CNN-based detectors suffer from enormous computational cost, which hinders them from real-time inference in computation-constrained scenarios. In this paper, we investigate the effectiveness of two-stage detectors in real-time generic detection and propose a lightweight twostage detector named ThunderNet. In the backbone part, we analyze the drawbacks in previous lightweight backbones and present a lightweight backbone designed for object detection. In the detection part, we exploit an extremely efficient RPN and detection head design. To generate more discriminative feature representation, we design two efficient architecture blocks, Context Enhancement Module and Spatial Attention Module. At last, we investigate the balance between the input resolution, the backbone, and the detection head. Compared with lightweight one-stage detectors, ThunderNet achieves superior performance with only 40% of the computational cost on PASCAL VOC and COCO benchmarks. Without bells and whistles, our model runs at 24.1 fps on an ARM-based device. To the best of our knowledge, this is the first real-time detector reported on ARM platforms. Code will be released for paper reproduction.
26
star
2

Stereo-OF-VO

Containing a wrapper for libviso2, a visual odometry library. The project about Optical flow and ORB and Libviso = visual odometry
C++
22
star
3

PPP-RTK

SPP、RTD、PPP、RTK、PPP-RTK、RAIM、ARAIM et al
C
14
star
4

ImageProcessing_Mathematical_Modeling

“华为杯”第十四届中国研究生数学建模竞赛（图像处理篇）——获全国二等奖(基于监控视频的前景目标提取)
MATLAB
12
star
5

RTCM3.3

RTCM3.3
12
star
6

Learning_TensorFlow-Kaggle_MNIST

一步步带你通过项目（MNIST手写识别）学习入门TensorFlow以及神经网络的知识
Python
11
star
7

RAIM

RAIM（Receiver Autonomous Integrity Monitoring）
MATLAB
9
star
8

TensorMask-Review

Sliding-window object detectors that generate boundingbox object predictions over a dense, regular grid have advanced rapidly and proven popular. In contrast, modern instance segmentation approaches are dominated by methods that first detect object bounding boxes, and then crop and segment these regions, as popularized by Mask R-CNN. In this work, we investigate the paradigm of dense slidingwindow instance segmentation, which is surprisingly underexplored. Our core observation is that this task is fundamentally different than other dense prediction tasks such as semantic segmentation or bounding-box object detection, as the output at every spatial location is itself a geometric structure with its own spatial dimensions. To formalize this, we treat dense instance segmentation as a prediction task over 4D tensors and present a general framework called TensorMask that explicitly captures this geometry and enables novel operators on 4D tensors. We demonstrate that the tensor view leads to large gains over baselines that ignore this structure, and leads to results comparable to Mask R-CNN. These promising results suggest that TensorMask can serve as a foundation for novel advances in dense mask prediction and a more complete understanding of the task. Code will be made available.
9
star
9

RAIM_PANG_NAV

RAIM for PANG NAV a tool for processing GNSS measurements in SPP, including RAIM functionality
MATLAB
7
star
10

MPU6050_Kalman_PWM_remote

MPU6050 Kalman PWM_remote
C++
7
star
11

SLAM-for-Paper

SLAM related papers and mathematical materials
6
star
12

SLAM-for-Matlab

SLAM related matlab
MATLAB
6
star
13

Matlab-From-Zero-To-One

MATLAB quickstart
MATLAB
5
star
14

Optical_Flow_Based_on_Duality_Method

The three elements of optical flow are as follows: optical flow produces velocity field; it is a carrier with information and optical properties such as pixels; it is a projection imaging of 3D scene motion to two-dimensional plane. It is the instantaneous velocity field generated by the motion of pixels with grayscale on the plane of the image.
MATLAB
5
star
15

C_plus_Plus_From_Zero_To_One

C_plus_Plus_From_Zero_To_One, Here are some exercises about C++, From Netease cloud class to Lao Jiu Jun
C++
4
star
16

Python-CNN-material

Python、神经网络相关资料
Jupyter Notebook
4
star
17

VisualOdometry_BasedOnSURF

%% Estimating the pose of the second view relative to the first view %% Bootstrapping estimating camera trajectory using global bundle adjustment %% Estimating remaining camera trajectory using windowed bundle adjustment
MATLAB
3
star
18

Data_Structures_and_Algorithms

Data Structures and Algorithms with C program
C
2
star
19

ORBSLAM2

2
star
20

rtklibdemo5

C
1
star
21

pandas_Python_Data_crawler

This is my first pandas-based web crawler project
Jupyter Notebook
1
star