Reading 10 : Visual Similarity of Pen Gestures

Citation:

Long Jr, A. Chris, et al. "Visual similarity of pen gestures." Proceedings of the SIGCHI conference on Human Factors in Computing Systems. ACM, 2000.

Publication Link

Summary:

This paper deals with an investigation into gesture similarity, and explains the methods used to come up with a set of the best 22 features to uniquely identify a gesture in feature space, by making use of the notion of perceptual similarity.

Discussion:

Based on experiments, it was found that the logarithm of quantitative correlates with similarity. It makes uses of MDS to translate the perceived distances between gestures, onto a multi-dimensional space. 

What exactly is MDS?

MDS pictures the structure of a set of objects from data that approximate the distances between pairs of the objects. The data points are arranged in this space so that the distances between pairs of points have the strongest possible relation to the similarities among the pairs of objects. That is, two similar objects are represented by two points that are close together, and two dissimilar objects are represented by two points that are far apart. From a slightly more technical point of view, what MDS does is find a set of vectors in p-dimensional space such that the matrix of euclidean distances among them corresponds as closely as possible to some function of the input matrix according to a criterion function called stress. Normally, MDS is used to provide a visual representation of a complex set of relationships that can be scanned at a glance. Since maps on paper are two-dimensional objects, this translates technically to finding an optimal configuration of points in 2-dimensional space. However, the best possible configuration in two dimensions may be a very poor, highly distorted, representation of your data. If so, this will be reflected in a high stress value. When this happens, you have two choices: you can either abandon MDS as a method of representing your data, or you can increase the number of dimensions. 

Experiment 1 : 

Participants were presented all possible triads of 14 gestures, and asked to identify the most different one in each triad. These response were recorded and use to compute a distance matrix between gestures.  The goals of this experiment were to identify the optimal number of features needed to represent the gesture, and to produce a model of gesture similarity.

Experiment 2:

3 more sets of 9 gestures each were added, and use evaluate the relationship between independent features.

Bottom Line : This is a very poorly written paper, and I really need to read this again to actually understand it.