Vision Based Natural Smart Phone Interface using Face Detection for Optimal UX

다국어 초록 (Multilingual Abstract)

User interface is the key technology in high-value smart phone applications services. This paper suggests a user interface algorithm and tests its validity through experiments. User’s face is detected in the input image on the smartphone camera sensor. Based on the sensing information about the detected face area, analysis is carried out to infer the locations for the eyes and the mouth. With the resulting information, the nose is detected. To identify the movements of each area, image-processing methods are applied to the eyes, the nose and the mouth. An algorithm is proposed to use them as interfaces. To test the effectiveness of the proposed algorithm, experiments are conducted on interfaces made with the proposed algorithm in the 3D augmented reality game environment with four virtual objects augmented. The test results show that the proposed algorithm is applicable for possible interfaces for various interactions with virtual objects in the augmented reality environment. This means that by realizing specialized interfaces on various smart phone services, exclusive conveniences are available satisfying critical and independent needs of the user. The proposed algorithm can also be used in user interfaces for optimal UX in AR Game where vision-processing methods are involved.

목차 (Table of Contents)

• Abstract
• 1. Introduction
• 2. Augment Reality Interface
• 3. Detection of Face
• 4. Proposed Algorithm
• 5. Face Interface
• 6. Experiments
• 6.1. Experiment 1
• 6.2. Experiment 2
• 6. 3. Experiment 3
• 6.4. Experiment 4
• 6.5. Experiment 5
• 6.6. Experiment 6
• 7. Conclusion
• References