Open source SIFT feature library Opensift:an open-source SIFT Library

Opensift An open-source SIFT LibraryView Project Ongithub

The scale invariant Feature Transform (SIFT) are a method to detect distinctive, invariant image Feature points, which easi Ly can is matched between images to perform tasks such as object detection and recognition, or to compute geometrical tran Sformations between images. The Open-source SIFT Library available here's implemented in C using the OpenCV open-source Computer Vision Library and I Ncludes functions for computing SIFT features in images, matching SIFT features between images using Kd-trees, and Computi Ng geometrical image transforms from feature matches using RANSAC. The library also includes functionality to import and work with image features from both David Lowe ' s SIFT executable Andt He Oxford Vgg ' s affine covariant feature detectors. The images below depict some of this functionality.

SIFT features detected in and images

SIFT features matched between the both images and the transform computed from the matches using RANSAC.

DependenciesOpenCV (>= 2.1) gtk+2 (>= 2.9). ReferencesAn Open Source SIFT Library. R. Hess. ACM MM, 2010. Distinctive Image Features from Scale-invariant keypoints. D. Lowe. IJCV, 60 (2), 2004. acknowledgmentsPlease see the THANKS file under the distribution for a list of contributors. Many thanks to all of these folks.
from:http://robwhess.github.io/opensift/

Patent Notice

The following patent have been issued for methods embodied in this software: ' Method and apparatus for identifying ' in Variant features in a image and use of same for locating a object in an image, "David G. Lowe, US patent 6,711,293 (Marc H 23, 2004). Provisional application filed March 8, 1999. Asignee:the University of British Columbia. For further details, contact David Lowe (lowe@cs.ubc.ca) or the University-industry liaison Office of the University of Br Itish Columbia