Evaluation of Sift and Surf for Vision Based Localization

Abstract

ABSTRACT:Vision based localization is widely investigated for the autonomous navigation and robotics. One of the basic steps of vision based localization is the extraction of interest points in images that are captured by the embedded camera. In this paper, SIFT and SURF extractors were chosen to evaluate their performance in localization. Four street view image sequences captured by a mobile mapping system, were used for the evaluation and both SIFT and SURF were tested on different image scales. Besides, the impact of the interest point distribution was also studied. We evaluated the performances from for aspects: repeatability, precision, accuracy and runtime. The local bundle adjustment method was applied to refine the pose parameters and the 3D coordinates of tie points. According to the results of our experiments, SIFT was more reliable than SURF. Apart from this, both the accuracy and the efficiency of localization can be improved if the distribution of feature points are well constrained for SIFT.

Evaluation of Several Feature Detectors/Extractors on Underwater Images towards vSLAM

“…Other evaluations are carried out for customized functions such as for tracking objects [18] and vision-based localization [19]. In [19] they added the Accelerated-KAZE (AKAZE) detector/descriptor to the review and analyzed the computing time.…”

“…Other evaluations are carried out for customized functions such as for tracking objects [18] and vision-based localization [19]. In [19] they added the Accelerated-KAZE (AKAZE) detector/descriptor to the review and analyzed the computing time. They also included Compute Unified Device Architecture (CUDA) implementations of AKAZE and SIFT being the fastest two in extracting, detecting and matching, followed by ORB and SURF.…”

“…Based on the performance of features extractors in the literature discussed in Section 1 we select SIFT [20], SURF [21], ORB [22], AKAZE [23] and BRISK [24] which are robust and have been used for indoor and outdoor environments in [19,25]. In Table 1 the characteristics of the detectors and descriptors are presented, as well as some parameters based on their OpenCV implementation.…”