Content uploaded by Mohammad Hafiz mypapit Ismail
Author content
All content in this area was uploaded by Mohammad Hafiz mypapit Ismail on Nov 25, 2014
Content may be subject to copyright.
2009 JITI
Journal of Information Technology Impact
Vol. 9, No. 3, pp. 173-182, 2009
Development and Evaluation of an Interactive 360° Virtual Tour for Tourist
Destinations
Aznoora Osman1 Nadia Abdul Wahab2 Mohammad Hafiz Ismail 3
Universiti Teknologi MARA
Malaysia
Abstract
This paper describes the development and evaluation of a virtual tour prototype for
tourist destinations. Four interesting destinations in Perlis, located in the northern
region of Malaysia were selected. It allows the users to navigate and to listen to the
background sounds and narration, as well as to read brief information about each place.
Each panorama contains hotspots that enable the users to further explore the
surroundings All pictures were combined using photo stitching technique to produce
panoramic images. The planar panorama was set as the panorama type.
The first
usability test was conducted to gain user feedbacks towards the virtual tour and to make
refinement to the prototype. The second usability test was intended to measure its
usability in terms of user satisfaction and effectiveness. The virtual tour received positive
feedbacks and all participants agreed that it could attract tourists to visit the actual
place.
Keywords: Planar panorama, photo-stitching, usability evaluation.
Introduction
Virtual reality (VR) can be defined as a three-dimensional multisensory, immersive, and
interactive digital environment that has triggered public imagination as future technology that will
dominate our work, education, and leisure (Roussou, 2004). On the other hand, virtual tour which
possesses virtual reality elements is virtual navigations of landscapes that exist in the real world
(Li, Lien, Chiu, & Yu, 1999). According to 360-virtual-tour-company.com, virtual tour is an
immersive technology that places the viewers inside the image, enabling them to significantly
enhance situational awareness and providing the highest level of functionality for viewing,
capturing and analyzing virtual data. Virtual tour can be clarified as a simulation of an existing
location that is composed of a sequence of video images. In addition, it can be developed from
photographic-based media or panoramas that have an unbroken view. All of these techniques
used to develop virtual tour were derived from virtual reality itself that is created by computer to
generate an experience of virtual world (Zhang, Lv, Zhang, Chen & Zhang, 2009). The virtual
tour developed in the research consists of still images that were combined using stitching
technique to produce panoramic images. The four destinations chosen were a snake farm, a
recreational lake, a cave and a waterfall. These places are among the most popular and mostly
visited by tourists when they come to Perlis.
Osman et al.
2009 JITI
174
Related Work
Wu et al. (2005) defined the three steps to realize tour in electric world whereby the first step is to
convert the entire real environment into electric models. The second step is to stitch the
photographs of actualities into a picture of virtual environment. Finally the third step is to mix the
previous two steps, by setting up the scene using 3D models combined with photograph
conjoining. However, the previous study made by Chen (1995) demonstrated a better technique
because it consisted of linking process that can link between the hotspots and the compression
steps in order to have a small size for fast loading.
In his study, Chen (1995) wrote on about the five steps in panoramic movie making, as
opposed to the three steps defined by Wu et.al (2005) which starts with node selection to maintain
the visual consistency when moving from one to another. The second step is stitching process to
create seamless panoramic image from a set of overlapping pictures. The third step is hot spot
marking which is to identify regions of panoramic image for interactions, such as navigation or
activating actions. The forth step is linking process where it connects and registers view
orientation between adjacent panoramic nodes. The last step is dicing and compression in which
the panoramic and hot spot images are diced before they are compressed and stored in a movie.
On the other hand, Li et al. (1999) focused on generating animations of customizable tour
paths and its associated human/camera motions. The system is capable in providing customizable
guided tours to a remote user. Besides that this system also incorporates motion-planning
algorithms to generate collision-free paths for the tour guide and for the virtual camera.
Improvements from Li et al. (1999) study were applied in a study of virtual tour into the
picture using a vanishing line and its extension to panoramic image that was done by Hyung et al.
(2001). They used spherical panoramic images that allow users to view from a camera which are
mapped onto the base sphere centered at the camera position. Finally they successfully shown
modeling scheme that naturally extended to navigation into a panoramic image. Furthermore they
provided the real sense of walk-through or navigation into the panoramic scene by enabling
continuous camera translation as well as rotation.
Besides studying the articles of previous virtual tour project, observation on technique used in
several virtual tours which were available through the Internet has also been made. Harvard
Virtual Tour that can be accessed from harvard.edu was designed and developed to allow people
around the world to take a look of the campus surroundings. It contains twenty places inside the
university that can be viewed virtually. They used Quick Time Virtual Reality (QTVR) as a
player to play the virtual tour. The virtual tour used cylindrical panorama in order to stitch images
and view the surroundings. It provides map for each location and a link to enter the hotspot, as
well as textual description of each place.
UiTM Virtual Tour that is embedded into UiTM Perlis website was developed to provide its
visitors with better snapshots of the campus. The images were stitched using spherical technique
and the virtual tour was displayed as cylindrical panoramas. This virtual tour provides four
hotspots that can be clicked. User can use its navigation tools to view in 360° rotation to zoom in
Virtual Tour for Tourist Destinations
JITI 2009
175
and to zoom out the surroundings. However, there was no narration or background sound
embedded into it.
In similar context, Rice University provides virtual tour with 300-acre campus view and its
urban surroundings that is embedded in its website located at www.rice.edu/virtualtours/. The
virtual tour was developed using high quality pictures and equipped with the university map to
allow users to click on the hotspots and further view the surrounding. This 360° degree
panoramic view put user in the centre of the university. The cubic panorama was adopted to
display the panoramic images. User can select an image to navigate around the places. Each
panorama was accompanied with brief description and useful links that give further insight into
why Rice is such a best place to study, work and play.
In contrast to other samples of virtual tours, virtual tour for University of Denver was
developed using sequence of video images. This virtual tour provided testimonials by the students
who were studying there. The video will change the scene depending on the student explanation.
User can easily choose another location if they wanted to change the scenery since the menu is
placed at the left side of the virtual tour itself. Nevertheless, the drawback is it does not provide
any navigation tool for the user. As such, user can only play, pause or stop the video.
On the other hand, the Mohave County virtual tour from virtualtourmedia.com provided more
choice of navigation where user can use keyboard to move around horizontally and vertically.
User can also zoom in or out while traversing the panorama. Since the photos used were high in
quality, the panorama remained clear and consistent while the user zoomed in. However the
virtual tour did not provide any clickable hotspot. Thus, user needs to choose another location
through hyperlinks that were displayed on the website.
Development
Techniques
Image stitching or photo stitching is the process of combining multiple photographic images with
overlapping fields of view to produce a segmented panorama or high-resolution image (Ward,
2006). The stitching quality is measured visually by the similarity of the stitched image to each of
the input images, and by the visibility of the seam between the stitched images (Zomet, Levin,
Peleg and Weiss, 2006). Nevertheless, researchers have proposed techniques for stitching
multiple photographs into single contiguous panoramic picture. Recent full view stitching
methods allow automatic picture stitching if they are taken in arbitrary order or spatial
arrangement (Baudisch, Tan, Steedly & Rudolph, 2005).
Software Used
The software packages used in developing this virtual tour application were Pixtra Panostitcher
v1.5, Easypano Tourweaver v5, Adobe Photoshop and Sony Sound Forge 8.0. Pixtra Panostitcher
was used to assist photos stitching process. It helped developer to stitch the images and also
allowed developer to manually adjust the overlap between photos during stitching process. Adobe
Photoshop was used to adjust the color balance and the brightness of the panorama. Moreover,
Osman et al.
2009 JITI
176
the process of editing and recording the sounds was done by using Sony Sound Forge software.
Finally, the Easypano Tourweaver was used to develop the application by importing the
panoramas, integrating the sounds and adding navigational features.
Steps
The prototype was developed by adapting the five steps defined by Chen (1995).
Node Selection
Node has been selected to maintain visual consistency when moving from one point to another.
The camera was mounted on a tripod and centered at its nodal point. Then the camera was rotated
in 360 degree of x-axis from the start point to the end point. Each image must have overlap with
the start point.
Photo Stitching
The purpose of stitching is to create an ideal panoramic image from a set of overlapping pictures.
In producing a good panoramic image, 50 percent overlap pictures is needed because the
adjoining pictures may have a very different brightness level (Chen, 1995). But it also may vary
depending on the image features in the overlapping regions. For a normal stitching session, the
pictures were stitched automatically and some pictures were stitched manually by adjusting some
points as remarks that were used to adjoin the pictures.
Image Compression
After stitching and editing process were completed, the panoramic image was resized to the
optimal size that can be rendered by the application. The image was compressed by reducing the
size from more than ten thousand pixels into 4996 pixels width.
Sound Recording and Editing
In this phase, the narration for each place was recorded and edited. Besides that, the background
sounds were also edited using the Sound Forge software. The editing was necessary to remove
noise that will affect the sound quality when it is integrated into the prototype.
Interface Design
Before developing the application, the template has been chosen from the software library. Some
hand drawn sketches were made to guide the design process. There were some adjustments made
to the template based on the sketches.
Virtual Tour for Tourist Destinations
JITI 2009
177
Sound Integration
Next phase is integrating the sounds into the virtual tour software. Sounds were imported into the
software after completely edited. The sound supported by software is in mp3 format. There were
four types of nature sounds that were imported which were sounds of bird, cricket, waterfall and
snake. Besides that, four sounds that contain the narration of each place were also imported.
Hotspots Marking
Hotspot (hot area) identifies regions of a panoramic image for interactions, such as navigation or activating
actions. The hotspot image does not need to have the same resolution as the panoramic image. By clicking
on the hot spot mark, this application will bring user to the hotspot region. Arrow symbols were elected as
indicator where this application will bring user to the hotspot areas once it is clicked.
Panoramic Linking
The linking process connects view orientation between adjacent panoramic nodes. The links were
attached to a hot spot so that the user may activate the link by clicking on the hotspot. In linking
process, besides creating a link between panoramic images, transition effect from the current
scene to the hot spot scene can also be selected. In addition, transition duration was set up in this
phase.
Usability Evaluation
The prototype was embedded into a university website and has undergone the first phase of
formative evaluation. The usability tests were conducted to gain user feedbacks and to measure
the usability of the prototype pertaining to user satisfaction and product effectiveness. The
usability tests were conducted in the computer laboratory of the university. One of the researchers
acted as the facilitator, who conducted the session. The participant was introduced to the
prototype and was briefed on the purpose of the usability test. In order to provide a comfortable
atmosphere during the test, the facilitator assured every participant that he or she was invited for
the purpose of evaluating the prototype only. Since it was found that testing with around five
users will discover most of usability problems in a particular interface (Nielson, 2000), the first
usability test was held with ten participants aged 20 to 40 years old; four males and six females.
The participants came from technical and non-technical background, and have never seen or used
a virtual tour before. However all of them have used the computer for more than three years.
They were four university students, two computing lecturers, one non-computing lecturers, two
Information Technology support officers and a clerk. All participants were required to interact
with the prototype according to a set of tasks.
The tasks centered on navigating the prototype such as exploring a destination, entering a
hotspot and viewing its surroundings as well as using the navigation tools and reading the textual
description. Only mouse was used as input device. After the usability test session, each
participant was informally interviewed regarding their opinion towards the prototype. The
feedbacks gathered from the test were used to make some refinement.
Osman et al.
2009 JITI
178
Then, the second usability test was performed with another five participants in the same age
range. They were a male general worker and four females who were a computing lecturer, a non-
computing lecturer, a preschool teacher and a primary school teacher. The participants were
required to perform the same tasks. Once they had accomplished all tasks, they were asked to fill
in an evaluation questionnaire to rate their satisfaction towards the speed of scene movement,
image quality, sounds and attractiveness of the virtual tour. In addition, they had to rate the
effectiveness of the application pertaining to the terminologies used, its navigability and the
textual description. The answers ranged from 1 (Strongly Disagree) to 4 (Strongly Agree).
Discussion and Findings
The first usability test elicited a few requirements and useful suggestions. They are summarized
in Table 1.
Table 1. Results of Usability Test 1
Evaluation Criteria Feedback
Speed of scene movement Moderate but when the navigation panel was
clicked, the speed became too fast.
Navigation Easy to navigate.
Background sounds Most participants liked the sounds. However, some
felt that the cricket sound were frightening.
Terminologies
Some terminologies were confusing such as there
were two terms of ‘Exit Cave’ in different
locations of a destination named Cave of
Darkness.
Quality of scenes/images Image quality was mostly satisfied by the
participants, but there should be a closed-up
photos of Snakes in the Snake Farm
Voice description Suitable and clear, but there were some uneven
tones.
Text description Clear and readable.
Attractiveness Yes, attractive.
Suggestions Add more attractive and famous places. Scene
lighting and color need to be enhanced.
Virtual Tour for Tourist Destinations
JITI 2009
179
Some refinements were done in term of sounds, button, speed, terminologies and transition
effects. The volume of cricket sound that was used as the background sound was reduced and
changed to a smoother sound of cricket to provide a more enjoyable experience while navigating.
Snake sound were edited to make it more similar to the real sound of snakes. From the
observation, none of the participants changed to full screen view since there was no button for it.
Initially, users had to right-click on the screen and choose full screen features. From the feedback,
one participant wanted to turn off the sound. Thus, the two buttons (full screen mode and sound
control) were added into the prototype. Furthermore, the problem in the speed of scene movement
was solved. The speeds for each panorama were decreased to three seconds per frame.
Terminologies that were used in the scene have been altered to more suitable terms that will guide
the user in their virtual visit. For example ‘Exit Cave’ that appeared twice in the scene was
changed to ‘Enter Cave’ and ‘Bath Area’ was changed to ‘Recreational Area’. Finally, the
transition effect such as slide, circle, spin and blinds were applied to all hotspots that were set for
each place.
In the second usability test, all evaluation criteria were either agreed or strongly agreed by the
participants. This indicated that the virtual tour received positive feedbacks from all participants.
Three participants agreed that the speed of scene movement and the background sound were
suitable. Four participants agreed that the images used were of high quality and that the virtual
tour could attract tourists to visit the actual place. In terms of the terminologies used, four
participants agreed that it was appropriate while three of them agreed that the textual description
in the application helped users to know the place better and that it was easy to navigate.
In general, all users were satisfied with the application. However, based on the score, there
are areas of improvement to make it more appealing. For example, it should be combined with
Flash animations, especially when the hotspots are clicked. More places should be added so that
tourists could get a glimpse of the interesting destinations in a particular state. Besides that
panoramic view of places such as the Snake Farm can be improvised by having a closed-up image
of snakes instead of image of snake cages and its surroundings. This application can also be
enhanced by showing the real activity that can be done at the selected places such as canoeing
activity by the lake area and integrating video of snake shows at the Snake Farm. Moreover, to
enhance the user experience while browsing through the application, it should be provided with a
3D walkthrough. Nonetheless, it was observed that users could easily adapt to the virtual tour and
enjoyed themselves while visiting the places virtually.
Conclusions and Recommendations
The research discovered user preferences towards a virtual tour for tourist destinations. It was
found that virtual tour can be used as a tool to promote tourism since it provides attractive
panoramic view as compared to still images found in brochures and websites. However, since the
target users come from all walks of life and varied in age range, more usability tests should be
done so that the end product will satisfy the requirements of most users.
Osman et al.
2009 JITI
180
References
Baudisch, P., Tan, D., Steedly, D. & Rudolph, E. (2005). Panoramic Viewfinder: Providing A
Real-Time Preview To Help Users Avoid Flaws In Panoramic Pictures. Proceedings of
OZCHI 2005, 1-10.
Chen, S. E. (1995). QuickTime VR-An Image Based Approach to Virtual Environment
Navigation. ACM SIGGRAPH International Conference on Computer Graphics and
Interactive Techniques, 29-38.
Harvard Virtual Tour. Retrieved April 1, 2009, from www.hno.harvard.edu/tour/
Hyung W. K., Soon, H. P., Ken-ichi, A. & Sung, Y. S. (2001). Tour Into the Picture using a
Vanishing Line and its Extension to Panoramic Images. EUROGRAPHICS2001, 20(3), 132-
141.
Nielsen, J. (2000). Why You Only Need to Test With 5 Users. Retrieved March 19, 2009, from
www.useit.com/alertbox/20000319.html
Pan, Z., et. al (2004). Easy Tour: A New Image-Based Virtual Tour System. ACM SIGGRAPH
International Conference, Session 8-3, (pp. 467-471).
Roussou, M. (2004). Learning by Doing and Learning Through Play: An Exploration of
Interactivity in Virtual Environments for Children. ACM Computers in Entertainment, 2(1),
1-23.
Schmidt N., Krone J. (2005). Constructing An Efficient And Easily Distributable Virtual Tour.
ACM Computing Literature, 3(1), 43-47.
Tourism Terengganu Virtul Tour. Retrieved September 23, 2009, from
http://tourism.terengganu.gov.my
UiTM Perlis Virtual Tour. Retrieved April 1, 2009, from www.perlis.uitm.edu.my
Virtual Tour of National Mall. Retrieved April 1, 2009, from http://dcpages.com/Tourism
Virtual Tour of Rice University. Retrieved September 9, 2009, from www.rice.edu/virtualtours
Ward, G. (2006). Hiding seams in high dynamic range panoramas. ACM International
Conference Proceeding Series, 153, p. 150.
Wu, S., Wang, R., & Wang, J. (2005). Campus Virtual Tour System based on Cylindric
Panorama. Department of Computer Science, Tsinghua University.
Zomet, A., Levin, A., Peleg, S. and Weiss, Y. (2006). Seamless image stitching by minimizing
false edges, IEEE Trans. Image Processing, 15(4), 969 – 977.
360° Virtual Tours. Retrieved March 28, 2009, from www.360-virtual-tour-company.com/360-
virtual-tour-samples.html
Virtual Tour for Tourist Destinations
JITI 2009
181
1 Aznoora Osman is a lecturer at the Faculty of Computer and Mathematical Sciences, Universiti Teknologi
MARA, Malaysia. She can be reached at Computer Science Department, Faculty of Computer and
Mathematical Sciences, Universiti Teknologi MARA Perlis, 02600 Arau, Perlis, Malaysia. Email:
aznoora@gmail.com; Phone: +6049882321
2 Nadia Abdul Wahab is a lecturer at the Faculty of Computer and Mathematical Sciences, Universiti
Teknologi MARA, Malaysia. She can be reached at Computer Science Department, Faculty of Computer
and Mathematical Sciences, Universiti Teknologi MARA Perlis, 02600 Arau, Perlis, Malaysia. Email:
nadiawahab@perlis.uitm.edu.my; Phone: +6049882745
3 M.Hafiz Ismail is a lecturer at the Faculty of Computer and Mathematical Sciences, Universiti Teknologi
MARA, Malaysia. He can be reached at Computer Science Department, Faculty of Computer and
Mathematical Sciences, Universiti Teknologi MARA Perlis, 02600 Arau, Perlis, Malaysia. Email:
mypapit@gmail.com; Phone: +6049882321
Osman et al.
2009 JITI
182
Page left blank
