In this paper we study interest point descriptors for image matching and 3D reconstruction. We examine the building blocks of descriptor algorithms and evaluate numerous combinations of components. Various published descriptors such as SIFT, GLOH, and Spin images can be cast into our framework. For each candidate algorithm we learn good choices for parameters using a training set consisting of patches from a multi-image 3D reconstruction where accurate ground-truth matches are known. The best descriptors were those with log polar histogramming regions and feature vectors constructed from rectified outputs of steerable quadrature filters. At a 95% detection rate these gave one third of the incorrect matches produced by SIFT.
|Published - Jun 2007
|CVPR '07: IEEE Conference on Computer Vision and Pattern Recognition, 2007 - Minneapolis
Duration: 17 Jun 2007 → 22 Jun 2007
|CVPR '07: IEEE Conference on Computer Vision and Pattern Recognition, 2007
|17/06/07 → 22/06/07