Gaze Interface for Mobile Device Operation

2015 Fiscal Year Final Research Report

• Project/Area Number: 25330189
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Perceptual information processing
• Research Institution: The University of Electro-Communications
• Principal Investigator: Takahashi Hiroki 電気通信大学, 情報理工学(系)研究科, 准教授 (80262286)
• Project Period (FY): 2013-04-01 – 2016-03-31
• Keywords: 視線計測 / 視線ジェスチャ / サッカード / 視線尤度 / HMM(Hidden Markov Model)

Outline of Final Research Achievements

This research develops a Gaze Interface for Mobile Devices Operation. (1) Eye gaze vectors are acquired with a single camera attached on a table PC. Although eye gaze vectors in the vertical direction sometimes include errors caused by an eyebrow or an eyelid, those in the horizontal direction are relatively measured correctly. (2) A gaze point is regarded to spread spatially and temporally. An icon selection method based on the distributions of gaze points in spatially and temporally is proposed. Target selections are succeeded extremely about 45% to 85%. (3) A gaze gesture classification method using HMM(Hidden Markov Model) that is capable to be used in low-end gaze tracking devices. Although the gesture classification performance strongly depends on users, it improves performance comparing with a rule-base classification method.

Free Research Field

総合領域