Impact Assessment of Image Feature Extractors on the Performance of Slam Systems

Abstract

This work evaluates an impact of image feature extractors on the performance of a visual SLAM method in terms of pose accuracy and computational requirements. In particular, the S-PTAM (Stereo Parallel Tracking and Mapping) method is considered as the visual SLAM framework for which both the feature detector and feature descriptor are parametrized. The evaluation was performed with a standard dataset with ground-truth information and six feature detectors and four descriptors. The presented results indicate that the combination of the GFTT detector and the BRIEF descriptor provides the best trade-off between the localization precision and computational requirements among the evaluated combinations of the detectors and descriptors.

DepthTiling: A novel way to increase visual SLAM performance in featureless environments

“…Features are distinctive points at the images and can be detected using any feature extraction technique. Yet, most of the visual systems uses combination of GFTT [1] and BRIEF [2] that is considered as most preferable [3]. Although various successful visual SLAM systems have been developed for last decades, it can be said that the state-of-the-art is RTAB-Map [4].…”