We present a new approach for selecting image pairs that are more likely to match in Structure from Motion (SfM). We propose to use Jaccard Similarity (JacS) which shows how many different visual words is shared by an image pair. In our method, the similarity between images is evaluated using JacS of bag-of-visual-words in addition to tf-idf (term frequency-inverse document frequency), which is popular for this purpose. To evaluate the efficiency of our method, we carry out experiments on our original datasets as well as on "Pantheon" dataset, which is derived from Flickr. The result of our method using both JacS and tf-idf is better than the results of a standard method using tf-idf only.
Selecting visually overlapping image pairs without any prior information is an essential task of large-scale structure from motion (SfM) pipelines. To address this problem, many state-of-the-art image retrieval systems adopt the idea of bag of visual words (BoVW) for computing image-pair similarity. In this paper, we present a method for improving the image pair selection using BoVW. Our method combines a conventional vector-based approach and a set-based approach. For the set similarity, we introduce a modified version of the Simpson (m-Simpson) coefficient. We show the advantage of this measure over three typical set similarity measures and demonstrate that the combination of vector similarity and the m-Simpson coefficient effectively reduces false positives and increases accuracy. To discuss the choice of vocabulary construction, we prepared both a sampled vocabulary on an evaluation dataset and a basic pretrained vocabulary on a training dataset. In addition, we tested our method on vocabularies of different sizes. Our experimental results show that the proposed method dramatically improves precision scores especially on the sampled vocabulary and performs better than the state-of-the-art methods that use pre-trained vocabularies. We further introduce a method to determine the k value of top-k relevant searches for each image and show that it obtains higher precision at the same recall.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.