Auto-framing: composition optimization combing subjective and objective measures

2015 Fiscal Year Final Research Report

• Project/Area Number: 26560006
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Multi-year Fund
• Research Field: Design science
• Research Institution: University of Yamanashi
• Principal Investigator: MAO Xiaoyang 山梨大学, 総合研究部, 教授 (20283195)
• Co-Investigator(Kenkyū-buntansha): TOYOURA Masahiro 山梨大学, 大学院総合研究部, 助教 (80550780)
• Project Period (FY): 2014-04-01 – 2016-03-31
• Keywords: オートフレーミング / 美的評価 / 注意獲得

Outline of Final Research Achievements

This research aims to develop an auto-framing technique enabling users to obtain the photographs of best composition. Taking a photograph of good composition requires skills and acknowleges. On the other hand, the measure to composition can be subjective. People may focus on different parts of an object or an scene. We found through the preliminary study that almost all users move camera before taking a photo due to the hesitation about the composition. The new auto-framing technique detects user’s subjective preference by analyzing the video sequence recorded during use's hesitation. Then Rule of Third is used as the objective aesthetic measure and the photograph of best composition is obtained through the optimization combing the subjective and objective measures.

Free Research Field

画像処理