A comprehensive evaluation of local detectors and descriptors

• Published in: Signal processing. Image communication 2017-11, Vol.59, p.150-167
• Main Authors: Wu, Song, Oerlemans, Ard, Bakker, Erwin M., Lew, Michael S.
• Format: Article
• Language: English
• Subjects: Computer vision; Computers; Detectors; Evaluation; Fully affine space; Image detection; Local descriptor; Local detector; Sensors; Studies; Visual words model
• ISSN: 0923-5965; 1879-2677
• DOI: 10.1016/j.image.2017.06.010

As local detectors and descriptors can find and represent distinctive keypoints in an image, various types of keypoints detection and description methods have been proposed. Each method has particular advantages and limitations and may be appropriate in different contexts. In this paper, we evaluate the performance of a wide set of local detectors and descriptors. First, we compare diverse local detectors with regard to the repeatability, and local descriptors in terms of the recall and precision. Next, we apply the visual words model constructed from the local descriptors with real values and binary string to large scale image search. The evaluation results reveal some strengths and weaknesses of the recent binary string descriptors compared with the notable real valued descriptors. Finally, we integrate the local detectors and descriptors with the framework of fully affine space and evaluate their performance under major viewpoint transformations. The presented comparative experimental studies can support researchers in choosing an appropriate local detector and descriptor for their specific computer vision applications. •First, the repeatability performance and the computational cost of each local detector are presented.•Additionally, the efficiency and accuracy of both the real valued descriptors and binary string descriptors in terms of recall and precision are evaluated.•Second, the visual words constructed from the real valued descriptors and binary string descriptors are evaluated for the application of large scale image search.•Third, we calculate the accuracy and time complexity of each local detector and descriptor in the framework of fully affine space such that researchers could make a trade-off between precision and efficiency under extreme viewpoint changes.