상세정보

Recently, multitasking on smartphones during navigation has emerged as a problematic social behavior due to its potential danger.  This thesis proposes to alleviate the situation by employing a sensor system for obstacle detection and aid the multitasking user so that one can safely navigate and carry out the on-going secondary task as effectively as possible.  The experiments have shown that with the aid of the sensor system, the user’s attention switch was significantly reduced.  However, for the case of casual video watching while navigating, such a reduction did not immediately translate to safer navigation nor particularly higher content concentration even after an extended period of usage.  This is attributed to the power and innate human peripheral vision and the secondary and relatively easy nature of the video watching task.  Nevertheless, the reduced attention shifts indicate that the user would be able to carry out both tasks more effectively.
In addition, I have found that event notification (e.g. incoming obstacles) in such multitasking situation was critical to the user performance, and the alerts “augmented” onto the background application or contents often had readability problems.  The visibility and readability of the augmented information must be as high as possible amidst the dynamically changing real world surrounding and background to ensure the alert is noticed quickly and understood.  Therefore, in this thesis, I also present a technique based on the image saliency analysis to improve the conspicuity of the foreground augmentation to the background real world medium by adjusting the local brightness contrast.  The proposed technique is implemented on a mobile platform considering the usage nature of AR.  Experiments were carried out to validate that the proposed simplified technique still improved information visibility and readability without much perceptual difference to the “optimal” ground truth cases.  Furthermore, the most usable way to deploy such a technique on-line was investigated by comparing four different representative schemes.  The results suggested that the users clearly preferred the automatic contrast modulation upon large movements in the scenery.
Finally, the saliency based contrast modulated augmentation technique was integrated into the obstacle alert and navigation aid system, as called the SalientAlert.  The combined system also represents a typical mobile AR application as well.  A pilot test has informally confirmed the effectiveness of the saliency adjustment of the augmented alerts.  The work in this thesis can contribute to providing more visible, readable and understandable contents and services for the mobile AR applications in general.

ABSTRACT	i
CONTENTS	iii
List of Figures	vi
List of Tables	xi
Chapter 1.	Introduction	1
 1.1 Motivation	1
 1.2 Objectives and Contributions	5
 1.3 Overview of Thesis Structure	6
Chapter 2.	Related Work	7
 2.1 Interactions while Walking	7
 2.2 AR Visualizations	8
Chapter 3.	Sensor-based Collision Avoidance Alert System	11
 3.1 Smartphone based Obstacle Detection	11
 3.2 Experiment 1	12
  3.2.1 Hypothesis and Experimental Procedure	12
  3.2.2 Experimental Task and Design	13
  3.2.3 Detailed Experimental Procedure	13
  3.2.4 Results	16
 3.3 Experiment 2	18
  3.3.1 Hypothesis and Experimental Procedure	18
  3.3.2 Experimental Task and Design	18
  3.3.3 Detailed Experimental Procedure	20
  3.3.4 Results	21
 3.4 Discussion	23
Chapter 4.	Improving Visibility of Augmentations	25
 4.1 Brightness Modulation for Visibility	25
 4.2 Performance	29
 4.3 Experiment 3	32
  4.3.1 Experimental Design and Tasks	32
  4.3.2 Detailed Procedure	34
  4.3.3 Results	35
  4.3.4 Discussion	37
 4.4 Experiment 4	40
  4.4.1 Applying the Proposed Algorithm for AR	40
  4.4.2 Experimental Design and Tasks	41
  4.4.3 Experimental Setup	42
  4.4.4 Detailed Experimental Process	43
  4.4.5 Results	46
  4.4.6 Discussion	55
Chapter 5.	SalientAlert: Salient Collision Avoidance Alert for User Safety in Mobile AR Environment	58
 5.1 Collision Avoidance System with Saliency Adjustment	58
 5.2 Experiment 5	59
  5.2.1 Experimental Design and Tasks	59
  5.2.2 Detailed Experimental Procedure	60
  5.2.3 Results	61
  5.2.4 Discussion	65
Chapter 6.	Conclusions and Future Work	67
 6.1 Conclusions	67
 6.2 Limitations and Future Work	67
References	69