Books:
"Visual Odometry: Part I - The First 30 Years and Fundamentals" by Davide Scaramuzza and Friedrich Fraundorfer
This is a comprehensive overview of the methods and challenges in visual odometry. It's a great starting point for anyone new to the topic.
"State Estimation for Robotics" by Timothy D. Barfoot
While not exclusively about visual odometry, this book provides a deep dive into state estimation techniques, which are crucial for VO.
Papers and Articles:
"ORB-SLAM2: An Open-Source SLAM System for Monocular, Stereo, and RGB-D Cameras" by Raúl Mur-Artal and Juan D. Tardós
A state-of-the-art SLAM system that incorporates visual odometry. The paper provides insights into the algorithms and techniques used.
"SVO: Fast Semi-Direct Monocular Visual Odometry" by Christian Forster, Matia Pizzoli, and Davide Scaramuzza
This paper introduces a semi-direct VO algorithm that's efficient and robust.
"DSO: Direct Sparse Odometry" by Jakob Engel, Vladlen Koltun, and Daniel Cremers
A novel direct sparse method for visual odometry that offers high accuracy.
Online Resources:
Visual Odometry Tutorials by Avi Singh
A series of tutorials that cover the basics of visual odometry, including feature detection, pose estimation, and optimization
OpenCV Tutorials
OpenCV, a popular computer vision library, has tutorials and functions related to camera calibration, stereo correspondence, and epipolar geometry, which are foundational for VO.
Robotics: Perception (Coursera)
This course by the University of Pennsylvania provides a good introduction to visual perception in robotics, including visual odometry.
Leveraging Deep Learning for Visual Odometry Using Optical Flow
This paper discusses the use of deep learning approaches for monocular visual odometry. It highlights the effectiveness of deep learning solutions in VO applications.
Recurrent Neural Network for (Un-)Supervised Learning of Monocular Video Visual Odometry
This work proposes a framework that simultaneously estimates visual odometry and depth maps from a video sequence taken by a monocular camera.
Unsupervised Collaborative Learning of Keyframe Detection and Visual Odometry
This paper focuses on keyframe selection in visual SLAM, which aids in efficient camera re-localization and the augmentation of visual odometry.
Visual Odometry Part I and Part II (Tutorial)
A two-part tutorial series offered by the Technical University of Munich. It provides a comprehensive introduction to visual odometry, including both theoretical concepts and practical exercises.
VisualSFM: A Visual Structure from Motion System
VisualSFM is a GUI application for 3D reconstruction using structure from motion (SfM). The website provides tutorials and resources related to the software, which can be a practical introduction to some concepts of visual odometry.
Robot Operating System (ROS) Tutorials
While ROS is a broad robotics middleware, it has packages and tutorials specifically related to visual odometry. If you're interested in robotic applications, exploring ROS can be beneficial.
YouTube Tutorials
There are several YouTube channels dedicated to computer vision and robotics. Searching for "visual odometry tutorials" on YouTube can yield many video tutorials and lectures that can be helpful.
Software and Datasets:
ORB-SLAM2 GitHub Repository
An open-source implementation of the ORB-SLAM2 system. It's a great resource for hands-on experimentation.
TUM MonoVO Dataset
A dataset from the Technical University of Munich (TUM) designed specifically for monocular visual odometry.
KITTI Vision Benchmark Suite
A comprehensive dataset with various benchmarks, including visual odometry. It's widely used in the research community.