Performance evaluation of single and cross‐dimensional feature detection and description

Abstract

Three-dimensional local feature detection and description techniques are widely used for object registration and recognition applications. Although several evaluations of 3D local feature detection and description methods have already been published, these are constrained in a single dimensional scheme, i.e. either 3D or 2D methods that are applied onto multiple projections of the 3D data. However, cross-dimensional (mixed 2D and 3D) feature detection and description has yet to be investigated. Here, we evaluated the performance of both single and cross-dimensional feature detection and description methods on several 3D datasets and demonstrated the superiority of cross-dimensional over single-dimensional schemes. 2D / 3D Keypoint Detection and Feature Description Methods Keypoint DetectorsKeypoint detectors analyse the structure around a vertex or a pixel depending on the data domain (3D or 2D,

Two-Fold and Symmetric Repeatability Rates for Comparing Keypoint燚etectors

“…where N X rp (d) is the number of repeated keypoints between the two images in the domain of image I x , and N ref is the number of keypoints in the reference image that appear in the common area. The same definition has recently been applied to 3D datasets [20] and far-infrared and thermal images [21]. In [22], the authors used a similar criterion but divided by a fixed number of keypoints.…”