Nearest Neighbors Distance Ratio Open-Set Classifier

<p>Feature vectors used in "Nearest neighbors distance ratio open-set classifier" paper to appear in Springer Machine Learning journal.</p> <p>In the <strong>15-Scenes</strong> (15scenes.dat) dataset, with 15 classes, the 4,485 images were represented by a bag-of-visual-word vector created with soft assignment and max pooling, based on a codebook of 1,000 Scale Invariant Feature Transform (SIFT) codewords.</p> <p>The 26 classes of the <strong>letter</strong> (letter.dat) dataset represent the letters of the English alphabet (black-and-white rectangular pixel displays). The 20,000 samples contain 16 attributes.</p> <p>The <strong>Auslan</strong> (auslan.dat) dataset contains 95 classes of Australian Sign Language (Auslan) signs collected from a volunteer native Auslan signer. Data was acquired using two Fifth Dimension Technologies (5DT) gloves hardware and two Ascension Flock-of-Birds magnetic position trackers. There are 146,949 samples represented with 22 features (<em>x</em>, <em>y</em>, <em>z</em> positions, bend measures, etc).</p> <p>The <strong>Caltech-256</strong> (caltech256.dat) dataset comprises 256 object classes. The feature vectors consider a bag-of-visual-words characterization approach and contain 1,000 features, acquired with dense sampling, SIFT descriptor for the points of interest, hard assignment, and average pooling. In total, there are 29,780 samples.</p> <p>The <strong>ALOI</strong> (aloi.dat) dataset has 1,000 classes and 108 samples for each class (108,000 in total). The features were extracted with the Border/Interior (BIC) descriptor and contain 128 dimensions.</p> <p>The <strong>ukbench</strong> (ukbench.dat) dataset comprises 2,550 classes of four images each. In our work, the images were represented with BIC descriptor (128 dimensions).</p>