Visual Odometry David Nister, CVPR 2004

Slides:



Advertisements
Similar presentations
3D Model Matching with Viewpoint-Invariant Patches(VIP) Reporter :鄒嘉恆 Date : 10/06/2009.
Advertisements

The fundamental matrix F
Registration for Robotics Kurt Konolige Willow Garage Stanford University Patrick Mihelich JD Chen James Bowman Helen Oleynikova Freiburg TORO group: Giorgio.
For Internal Use Only. © CT T IN EM. All rights reserved. 3D Reconstruction Using Aerial Images A Dense Structure from Motion pipeline Ramakrishna Vedantam.
Summary of Friday A homography transforms one 3d plane to another 3d plane, under perspective projections. Those planes can be camera imaging planes or.
MASKS © 2004 Invitation to 3D vision Lecture 7 Step-by-Step Model Buidling.
Recent work in image-based rendering from unstructured image collections and remaining challenges Sudipta N. Sinha Microsoft Research, Redmond, USA.
Object Recognition using Invariant Local Features Applications l Mobile robots, driver assistance l Cell phone location or object recognition l Panoramas,
Computer vision: models, learning and inference
Multiple View Reconstruction Class 24 Multiple View Geometry Comp Marc Pollefeys.
Hybrid Position-Based Visual Servoing
Parallel Tracking and Mapping for Small AR Workspaces Vision Seminar
Robust Object Tracking via Sparsity-based Collaborative Model
A Generic Concept for Camera Calibration Peter Sturm and Srikumar Ramaligam Sung Huh CPSC 643 Individual Presentation 4 April 15, 2009.
MASKS © 2004 Invitation to 3D vision Lecture 11 Vision-based Landing of an Unmanned Air Vehicle.
Adam Rachmielowski 615 Project: Real-time monocular vision-based SLAM.
Visual Odometry Michael Adams CS 223B Problem: Measure trajectory of a mobile platform using visual data Mobile Platform (Car) Calibrated Camera.
Feature matching and tracking Class 5 Read Section 4.1 of course notes Read Shi and Tomasi’s paper on.
Automatic Image Alignment (feature-based) : Computational Photography Alexei Efros, CMU, Fall 2005 with a lot of slides stolen from Steve Seitz and.
Feature tracking Class 5 Read Section 4.1 of course notes Read Shi and Tomasi’s paper on good features.
Visual Odometry for Ground Vehicle Applications David Nister, Oleg Naroditsky, James Bergen Sarnoff Corporation, CN5300 Princeton, NJ CPSC 643, Presentation.
Multiple View Geometry Marc Pollefeys University of North Carolina at Chapel Hill Modified by Philippos Mordohai.
Automatic Image Alignment (feature-based) : Computational Photography Alexei Efros, CMU, Fall 2006 with a lot of slides stolen from Steve Seitz and.
Sebastian Thrun CS223B Computer Vision, Winter Stanford CS223B Computer Vision, Winter 2005 Lecture 3 Advanced Features Sebastian Thrun, Stanford.
CSCE 641 Computer Graphics: Image-based Modeling (Cont.) Jinxiang Chai.
Visual Odometry Chris Moore Mark Huetsch Firouzeh Jalilian.
REAL-TIME DETECTION AND TRACKING FOR AUGMENTED REALITY ON MOBILE PHONES Daniel Wagner, Member, IEEE, Gerhard Reitmayr, Member, IEEE, Alessandro Mulloni,
55:148 Digital Image Processing Chapter 11 3D Vision, Geometry Topics: Basics of projective geometry Points and hyperplanes in projective space Homography.
The 2009 IEEE/RSJ International Conference on Intelligent Robots and Systems October 11-15, 2009 St. Louis, USA.
Automatic Camera Calibration
Computer vision: models, learning and inference
Kalman filter and SLAM problem
Factor Graphs Young Ki Baik Computer Vision Lab. Seoul National University.
Lecture 11 Stereo Reconstruction I Lecture 11 Stereo Reconstruction I Mata kuliah: T Computer Vision Tahun: 2010.
Final Exam Review CS485/685 Computer Vision Prof. Bebis.
SLAM (Simultaneously Localization and Mapping)
A General Framework for Tracking Multiple People from a Moving Camera
3D SLAM for Omni-directional Camera
Vision-based Landing of an Unmanned Air Vehicle
Flow Separation for Fast and Robust Stereo Odometry [ICRA 2009]
PRE-DECISIONAL DRAFT: For Planning and Discussion Purposes Only Test Plan Review MSL Focused Technology Instrument Placement Validation Test Plan for 2D/3D.
Dynamic 3D Scene Analysis from a Moving Vehicle Young Ki Baik (CV Lab.) (Wed)
Scientific Writing Abstract Writing. Why ? Most important part of the paper Number of Readers ! Make people read your work. Sell your work. Make your.
Frame Decimation for Structure and Motion Young Ki Baik CV Lab.
CSCE 643 Computer Vision: Structure from Motion
Correspondence-Free Determination of the Affine Fundamental Matrix (Tue) Young Ki Baik, Computer Vision Lab.
Young Ki Baik, Computer Vision Lab.
Computer Vision Lab Seoul National University Keyframe-Based Real-Time Camera Tracking Young Ki BAIK Vision seminar : Mar Computer Vision Lab.
3D Reconstruction Jeff Boody. Goals ● Reconstruct 3D models from a sequence of at least two images ● No prior knowledge of the camera or scene ● Use the.
Acquiring 3D models of objects via a robotic stereo head David Virasinghe Department of Computer Science University of Adelaide Supervisors: Mike Brooks.
Asian Institute of Technology
HONGIK UNIVERSITY School of Radio Science & Communication Engineering Visual Information Processing Lab Hong-Ik University School of Radio Science & Communication.
MTP FY03 Year End Review – Oct 20-24, Visual Odometry Yang Cheng Machine Vision Group Section 348 Phone:
Real-Time Simultaneous Localization and Mapping with a Single Camera (Mono SLAM) Young Ki Baik Computer Vision Lab. Seoul National University.
HONGIK UNIVERSITY School of Radio Science & Communication Engineering Visual Information Processing Lab Hong-Ik University School of Radio Science & Communication.
Fast Census Transform-based Stereo Algorithm using SSE2
3D reconstruction from uncalibrated images
Spatiotemporal Saliency Map of a Video Sequence in FPGA hardware David Boland Acknowledgements: Professor Peter Cheung Mr Yang Liu.
55:148 Digital Image Processing Chapter 11 3D Vision, Geometry Topics: Basics of projective geometry Points and hyperplanes in projective space Homography.
55:148 Digital Image Processing Chapter 11 3D Vision, Geometry Topics: Basics of projective geometry Points and hyperplanes in projective space Homography.
Vision-based SLAM Enhanced by Particle Swarm Optimization on the Euclidean Group Vision seminar : Dec Young Ki BAIK Computer Vision Lab.
Lecture 9 Feature Extraction and Motion Estimation Slides by: Michael Black Clark F. Olson Jean Ponce.
Visual Odometry for Ground Vehicle Applications David Nistér, Oleg Naroditsky, and James Bergen Sarnoff Corporation CN5300 Princeton, New Jersey
776 Computer Vision Jan-Michael Frahm Spring 2012.
1 Long-term image-based motion estimation Dennis Strelow and Sanjiv Singh.
MASKS © 2004 Invitation to 3D vision. MASKS © 2004 Invitation to 3D vision Lecture 1 Overview and Introduction.
1 Bilinear Classifiers for Visual Recognition Computational Vision Lab. University of California Irvine To be presented in NIPS 2009 Hamed Pirsiavash Deva.
55:148 Digital Image Processing Chapter 11 3D Vision, Geometry
Paper – Stephen Se, David Lowe, Jim Little
Features Readings All is Vanity, by C. Allan Gilbert,
Presentation transcript:

Visual Odometry David Nister, CVPR 2004 2005. 1. 4 Computer Vision Lab. Young Ki Baik

Contents Introduction Algorithm Experimental results Conclusion and opinion.

Introduction Visual Odometry Features for real-time Usage of visual information as a sensor Realization of the real-time navigation system using 3D reconstruction algorithms (camera motion estimation algorithm) Features for real-time Parallel processing based PC (MMX) Pentium III 1GHz Fast algorithm Preemptive RANSAC (ICCV2003) Features for accuracy Stereo camera Calibrated framework

Introduction System overview Feature extraction 3D reconstruction Matching and tracking Motion estimation 5-point algorithm / P-RANSAC Triangulation method Bundle adjustment Harris corner detector Normalized correlation 3-point algorithm for 3D motion

Algorithm Feature extraction Harris corner detector No subpixel precision detection Usage of down sampled data (16 bit) Size of INT and FLOAT is 32 bit. Low size of data can be expected more efficiency for parallel processing. 32 bit MMX register 16 bit 64bit

Algorithm Feature matching … … … Normalized correlation over an 11x11 window 11x11 = 121 (for applying to 128 bit aligned memory) Matching with converted 1 dimensional vector using Parallel processing (MMX) is faster than normal method. Short search range (Video sequences have short base line) 121 7 Garbage space … Matching using MMX … …

Algorithm 3D reconstruction 5-point algorithm Only considering pose estimation. Usage of 2D points. Preemptive RANSAC (CVPR 2003) Fast RANSAC Triangulation method Conventional triangulation method is used for 3D reconstruction. Bundle adjustment Using small number of parameters and iteration.

Algorithm Motion estimation R, T 3-point algorithm Only considering camera pose (rotation and translation) estimation. Usage of 3D point. Generated points R, T Triangle Selected points

Algorithm Merit of using the Stereo Vision Known scale (baseline) Less affection by uncertainty in depth

R, T Algorithm The Stereo Scheme Triangulation Stereo camera Matching 3D motion (3-P algo., P-RANSAC) Matching Next frame Motion estimation (5-P algo., P-RANSAC) R, T Stereo camera Matching Triangulation

Algorithm The Stereo Scheme Firewall Optimization (LM) 3D motion estimation Certain number of frames Optimization (LM) Coordinate system is transferred. Firewall For stopping propagation error

Experimental results System configuration Experiments Environment CPU : Pentium III 1GHz (MMX) Stereo camera (360*240*2) size / FOV : 50˚ / Baseline : 28 cm Experiments GPS : Location error test INS : Direction error test Environment Loop Meadow Woods

Experimental results Processing time Location error Direction error Around 13Hz Location error Direction error

Experimental results Performance Red line : Visual odometry Blue line : DGPS

Experimental results Performance Red : Visual odometry Blue : DGPS

Conclusion and Opinion. Real-time navigation system is implemented. Opinion There is no refinement scheme for solving closing loop problem. More fast result with Pentium-IV (SSE2) There is room for improvement.