International Journal of Computer Theory and Engineering

Abstract—This paper proposes a method for estimating the meanings for groups of shapes in presentation slides. The method of estimation is as follows; 1) Extracting features of each shape in a presentation slide, such as size, position, and type (e.g. circle, rectangle, arrow, etc.), 2) Analyzing the relations between these shapes, such as connection, inclusion, overlap, and similarity, 3) Grouping the shapes into meaningful groups using these relations, and estimating the meaning for each group, such as list, set, and flow/transition. The result of the proposed method could be used for retrieval and reuse of figures in presentation slides. In order to verify the effectiveness of the proposed method, we conducted an experiment of estimation accuracy, and achieved 0.93 in recall and 0.70 in precision.

Index Terms—PowerPoint, OpenXML, figure retrieval.

Y. Sakuragi is with at the Graduate School of Information Science and Engineering, Ritsumeikan University, 1-1-1 Noji-higashi, Kusatsu, Shiga 525-8577, Japan (e-mail: is0018ke@ ed.ritsumei.ac.jp).
A. Aoyama is with Graduate School of Technology Management, Ritsumeikan University, 1-1-1 Noji-higashi, Kusatsu, Shiga 525-8577, Japan (e-mail: aoyama@mot.ritsumei.ac.jp).
F. Kimura is with Kinugasa Research Organization, Ritsumeikan University, 56-1 Toji-in Kita-machi, Kita-ku, Kyoto, Kyoto 603-8577, Japan (e-mail: fkimura@ ed.ritsumei.ac.jp).
A. Maeda is with College of Information Science and Engineering, Ritsumeikan University, 1-1-1 Noji-higashi, Kusatsu, Shiga 525-8577, Japan (e-mail: amaeda@is.ritsumei.ac.jp).

[PDF]

Cite:Yuki Sakuragi, Atsushi Aoyama, Fuminori Kimura, and Akira Maeda, "A Method for Estimating Meanings for Groups of Shapes in Presentation Slides," International Journal of Computer Theory and Engineering vol. 8, no. 1, pp. 74-79, 2016.