Paper Review #5: Visual Similarity of Pen Gestures

1. Paper Bibliography

• Title: Visual Similarity of Pen Gestures
• Authors: A. Chris Long, Jr., James A. Landay, Lawrence A. Rowe, and Joseph Michiels
• Publication: Proceedings of the SIGCHI conference on Human factors in computing systems, ACM, 2000.

2. Summary

• This paper studied perceptual similarity of pen-based gestures. They had done two similarity tests with multiple-dimensional scaling(MDS). With MDS, they selected the features that affect gestures mostly. The similarity of the gestures is given by the Euclidean distance between the two gestures in the feature space, where smaller distance means greater similarity. The first trial got 0.74 of correlation. In the second trial, the computation model they studied predicted perceived gesture similarity that correlates 0.56 with observation.

3. My opinion

    1) Method

• They used Euclidean distance and MDS for their experiments. The paper said that depending on researches, researchers can use either Euclidean distance or Manhattan distance, and they got a better result with Euclidean distance, so they used it. I think this methodology is interesting.  

    2) Idea

• This idea gave me a lot of ideas for my research. I also need to do user study in order to get intuitive hand gestures which imply emotions.

    3) New research idea

• I think that the main steps of this experiment can be applied to any HCI user studies such as hand gestures.