Accomplished artists often include leading lines in their compositions, which utilize not only obvious lines but also rows of similar shapes, gradations, brush strokes, etc., to emphasize the main subject of their works, as well as to guide the viewer’s attention to the surrounding areas. Although these leading lines are hard to recognize for observers with no experience in art education, identifying these lines helps to understand the structure of artwork, and is said to be important in art education. Another problem is that the artist’s eyes become accustomed to their work over a long period, and the leading lines become hard to identify.
Saliency maps are used to locate attractive areas of still images and have been utilized in various fields, including advertising design, media quality assessment, and analysis of medical images. However, such scalar features do not express the trajectory of a viewer’s gaze.
In this project, we propose a visual analysis method for leading lines based on subgraphs of Morse-Smale complexes extracted from saliency maps. By abstracting the gradient field of saliency maps, we propose a method to identify leading lines for beginner artists and those who wish to learn how to appreciate art.
Members with gray background are not currently involved in the project. Members who still belong to Fujishiro Lab are marked with the Lab icon before their names.
|Fuminori Shibasaki||Keio University|
The underlined authors are current or past members of Fujishiro Lab.
- Fuminori Shibasaki and Issei Fujishiro: “A comparative study on visual analysis of leading lines based on differential topology and eye tracking results,” in Proceedings of Expressive Japan 2023 (The Technical Report of the Institute of Image Information and Television Engineers, Vol. 47, No. 9, pp.89―92, AIT2023–59), Tokyo Polytechnic University, online & onsite hybrid, March 6, 2023, Excellent Oral Presentation Awards, Polyphony Digital Award (in Japanese).
- Fuminori Shibasaki and Issei Fujishiro: “Visual analysis of gaze direction strategies on static images based on differential topology,” in Proceedings of the 85th National Convention of International Processing Society of Japan, Vol. 2, pp. 939―940 (7X-01), The University of Electro-Communications, online & onsite hybrid, March 2―4, 2023, Student Encouragement Awards (in Japanese).
- Grant-in-Aid for Challenging Research (Pioneering): 20K20481 (2022―)