GitHub - wired777b/tom-cv: Various computer vision routines by Tom Botterill. Essential matrix estimation+optimisation, BoWSLAM, BaySAC, real-time local mosaicing.

wired777b / tom-cv Public

forked from kayak-tom/tom-cv

Notifications You must be signed in to change notification settings
Fork 0
Star 0

Various computer vision routines by Tom Botterill. Essential matrix estimation+optimisation, BoWSLAM, BaySAC, real-time local mosaicing.

0 stars 20 forks Branches Tags Activity

Star

Notifications

Branches Tags

Name		Name	Last commit message	Last commit date
Latest commit History 2 Commits
BoWSLAM		BoWSLAM
BoW_Windows_Wrapper		BoW_Windows_Wrapper
ShowIm		ShowIm
bagOfWords		bagOfWords
bowExample		bowExample
bowSpeedometer		bowSpeedometer
cameraGeom		cameraGeom
cmake		cmake
featureDescription		featureDescription
featureExtract		featureExtract
featureMatching		featureMatching
fullModel		fullModel
guidedMatchingTest		guidedMatchingTest
hydrologicalModel		hydrologicalModel
image		image
imageSource		imageSource
logging		logging
mosaicing		mosaicing
params		params
ransac		ransac
ransacAndRefine		ransacAndRefine
ransacTest		ransacTest
timelapse		timelapse
timing		timing
util		util
CMakeLists.txt		CMakeLists.txt
CMakeLists.txt~		CMakeLists.txt~
Documentation.txt		Documentation.txt
License_and_Installation_Readme.txt		License_and_Installation_Readme.txt

Repository files navigation

LICENSE: 

Essential matrix estimation code (in 'ransac' library) is based on Henrik Stewenius's MATLAB code (http://vis.uky.edu/~stewe/FIVEPOINT/) and is for academic use only. Please see the notice in the file calibrated_fivepoint_helper.c.

All other code is free to use/modify/redistribute for any purpose. Please send me an email if you find it useful, or if anything doesn't work, or when you find bugs! tom.botterill - at - grcnz.com. 


PRE-REQUISITES:

Most components require:
1) boost (I use 1.39.0, works with other versions, libraries need to be built)
2) openCV (latest version 2.0 recomended, needed for homography estimation, N-point Essential matrix estimation (N>=8) and anything using images)
3) Eigen (a fast header-only matrix library, much faster if the convergence threshhold for eigenvector computation is raised and the complex eigenvector code is removed)


BUILD:

Project files for the following IDEs are included. You may have to fix some absolute paths, especially under Linux.

WINDOWS:
Visual Studio 2008 (open util.sln)
Visual Studio 2005 (MODIFY all the .sln and .vcproj files to change Version="9.00" to Version="7.00")

LINUX:
1) Make 3 symlinks in the workspace directory: ln -s PATH_TO_YOUR_OPENCV_DIR opencv; ln -s PATH_TO_YOUR_EIGEN_DIR opencv; ln -s PATH_TO_YOUR_BOOST_DIR boost

Eclipse CDT (Galileo)

The makefiles supplied are just the ones generated from Eclipse but should work. E.g. "cd util/Debug; make" will build libutil.a with debugging info, etc.


DOCUMENTATION:

Online at www.hilandtom.com/tombotterill/code


REFERENCING:

My publications (with links and bibtex) are listed at www.hilandtom.com/tombotterill

My Bag-of-Words scheme is described in: "Speeded-up Bag-of-Words algorithm for robot localisation through scene recognition" and used in "A Bag-of-Words Speedometer for Single Camera SLAM" and "An Integrated IMU, GNSS and Image Recognition Sensor for Pedestrian Navigation"

BaySAC is described in "New Conditional Sampling Strategies for Speeded-Up RANSAC"

Another paper on Visual SLAM (BoWSLAM) that uses all of this code is waiting for review; a video is online at www.hilandtom.com/tombotterill