A Survey on Feature Point Extraction Techniques

Feature point extraction plays a significant role in many computer vision tasks, such as visual simultaneous localization and mapping (VSLAM), structure from motion (SFM) and object recognition. The performance of feature point extraction is directly related to the accuracy of the final result. To obtain more repeatable feature points in the image, over the past decades, researchers have developed a large number of techniques to extract feature points which perform well in the challenging scenes. In this paper, we conduct a survey to introduce the development of feature point extraction. In addition, we also present a comparison of some famous techniques to show their performance on public datasets. Finally, we conclude the status of the feature point extraction and make an insightful discussion on the future direction.