Immersive Virtual Undersea World - onlinepresent.org

Advanced Science and Technology Letters Vol.39 (Games and Graphics 2013), pp.7-13 http://dx.doi.org/10.14257/astl.2013.39.02

Immersive Virtual Undersea World Jung-Yoon Kim1, Yoon-Seok Choi2, Soonchul Jung 3 Jin-Sung Choi4 and Won-Hyung Lee1* 1,1*

Graduate School of Advanced Imaging Science, Multimedia & Film, Chung-Ang University 2,3 Creative Content Research Laboratory, ETRI [email protected], [email protected], [email protected], [email protected], [email protected]

Abstract. This paper proposes a virtual undersea world for users based on the real world. As the sensor technology in which one can feel things as if they were in the real world owing to the development of equipment techniques like HMD and data gloves was introduced, real-time interactions with a virtual world is made possible. Based on this, this study generated images in the direction of the user experiencing the virtual world, and used the wireless HMD in a real world designed in a 1‐to‐1 size to the virtual world to support free exploration in the virtual world. Also, it enhanced the user’s sense of experience at the virtual undersea world by generating images with control of virtual fish in accordance with the user’s hand movements. Keywords: animation

1

virtual world, interaction, immersive HMD, reai-time crowd

Introduction

The virtual reality technology has been renewed repetitively while used in various fields including movies and games. In particular, equipment technologies like head mounted display (HMD) and data gloves have been developed so that the user can interact in real time with a virtual world by introducing technologies supporting vision, hearing, and touching in accordance with program control in order for the user to feel things as if he or she is in the real world when he wears HMD on his head and data gloves on his hands [1]. In particular, the realization of virtual objects is one of the main works in a virtual world in which the user is immersed in his own right, and feels his own existence to behave practically. This work is a course in which the shapes of the virtual objects are determined and the behavior of each object is expressed through the flow of time. The techniques of control of objects’ shape and animation of virtual objects’ movement are used to put life into the objects as in the real world so as to make the movements of the group objects with interaction of surrounding elements after modeling the motions of the virtual objects and the interactive motions among the objects. This paper propose a virtual undersea world in which the user can have an experience and be immersed, on the basis of studies on controlling natural motions of ISSN: 2287-1233 ASTL Copyright © 2013 SERSC

Advanced Science and Technology Letters Vol.39 (Games and Graphics 2013)

the virtual objects while vivifying their characteristics. In order to maximize the feelings of walking under the reality‐like sea, we represented a terrain of undeasea, various willowy seaweeds in tidal currents, and hundreds of fish filling up the seaweeds and rocks, and controlled their motions. Also, we supported interaction with the virtual world by sensing the user’s hand gestures with wireless HMD equipment and through marker‐based motion tracking. Then the paper is organized as follows. Section 2 introduces the virtual reality techniques and an experience system based on them. Section 3 accounts for the composition of the whole proposed system and basic methods of its representation. Section 4 concludes the paper.

2

Existing studies

2.1 Virtual Reality and crowd animation Much work has been carried out to effectively control crowd objects in a virtual world. As for the technique for expressing crwod animation, in 1987, the Reynolds proposed a particle system to simulate the movement of bird flocks with a bird assumed to be a particle [2]. In order to maximally reduce user input, the biggest problem with Reynolds’s study, Tu and Terzopoulos modeled a physics‐based virtual undersea world that can process every animal’s movement types just with a single program [3]. They defined artificial fish in the virtual undersea world as spring and mess physical models, generated fish movements through the length changes in spring of artificial fish, and constructed the system so as to make the fish automatically move naturally within the virtual undersea world by applying the artificial intelligence technology [3]. In order to represent virtual fish in this way, we have steering behavior, which expresses individual objects engaging in crwod movement. In order to realize artificial lives that can make us feel as if they are alive as in collective behavior, realistic animation techniques and a controlling structure showing intelligent behavioral modes are necessary. Boid, which was proposed by Reynolds, is a method to express emergence, the key feature of an artificial life, and is used to model collective behavior of animals, birds, and fish [2]. We regularize the Boids’ basic actions within a crwod into three types, separation, alignment, and cohesion, and call them steering behavior [Fig 1]. In addition to those three types of steering, this paper added and realized several natural fish’s behavior. Tarasewich used the concept of steering behavior for ants’ food search, fish’s movement, and birds’ movement in terms of the crwod behavior of individual objects that interact on the basis of local environment information [4].

8

Copyright © 2013 SERSC


Fig. 1. 3 rules: separation, alignment, and cohesion [2] What is important in VR is to naturally and effectively control users’ avatars in a virtual space so that they can interact with the virtual space and objects there. Interactions with a virtual space are classified into three types: manipulation, navigation, and communication [5]. Some systems began to provide user gestures. Takala et al. proposed a virtual aquarium system that adopted a gesture sensing system [6], in which users take a swimming motion to move forward in the virtual space. Such an experience of a virtual space in terms of real actions is simpler and more natural [7].

3

Immersive virtual undersea world

3.1 Configuration of the system In order to reflect users’ actual motions to virtual world, we preapared a 7m x 7m space in which motions can be tracked with a motion tracking device in a 10m square actual space. We used two servers, a motion tracking server and an image server, for the virtual undersea world. For motion tracking, we used OptiTrack S250e cameras and tracking tools in OptiTrack system produced by the Natrual Point Company to track user’s HMD motion and hand gestures. Based on the collected motion information, the image server determines the directions of user’s views, and generates images in the 720p resolution. Also, it senses users’ gestures by tracking markers attached to his hands. User can freely come about in the virtual undersea space because the generated user views are delivered to the user’s HMDs through an HDMI transmission device. [Fig. 2] is a scene from an images generated by the system.


9


Fig. 2. Virtual undersea world

3.2 Undersea scenes The system uses Ogre3D, an open game engine, for virtual undersea scenes rendering. Images for user are transmitted to the user in the way that information of translation or rotation that has been given by tracking the markers attached to the HMD devices equipped on user’s head is transmitted to the virtual aquarium system, and then it is generated as the images in real-time. In order to compose the virtual undersea world, we produced a total of 30 kinds of fish and 10 kinds of seaweeds. In order to enhance the quality, each fish model is basically assigned a normal map and a light map [Fig 2]. Also, in order to give undersea senses, GodRay effects and wave effects are added by applying GPU Shader, an optical refraction effect, and in order to show rich undersea spaces, more than 300 objects are rendered to be swimming underwater.

Fig. 2. Fish modeling process

10



As for fish school behavior, their positions are evaluated on the three rules of separation, cohesion, and alignment, and their movement speeds are calculated [2]. This algorithm includes the following problems. Computation costs increase in proportion with the number of objects, and in addition, there are cases where movement is impossible with the three rules. In particular, the Computation costs in proportion with proliferation of objects lower rendering speeds in scenes where users are surrounded by fish in school. To solve this problem, we added two measures. First, we used speed values resulting from schooling calculations to set up target positions, and lowered calculation frequencies by allowing only arrival at the established target positions to be calculated. Second, we added two rules, avoidance and path following. Collision between objects is maximally avoided under the rule of avoidance, while the phenomenon of interruption due to use of schooling computations in case of non‐movement is eliminated and fish are led to maximally follow established paths under the rule of path following. The virtual fish model has 4 states, ‘idle, eat, interaction, wander,’ and each state is defined as follows [Table 1]. Virtual fish take actions by shifting from a state to another in terms of gesture events or certain probabilities [Fig. 3]. Table 1. 4 States

State

Description

Idle

A basic state in which fish wait for an event for basic movement and shifting to another state.

Eat

A fish shifts to this state when it detects a prey in the state of Idle.

Interaction

This state is one that deals with events in terms of gestures.

wander

In this state, fish randomly swim about in a space.

Fig. 3. State transition diagram

Also, we have defined 10 gestures in order to allow users to interact with fish in the virtual undersea world. A user moves his left or right hand in some pre-defined direction, fish moves in the same direction. In addition, the system supports 3D‐streoscopic image generation, which is printed out to the HMD to enhance the user’s immerse in the virtual space. Copyright © 2013 SERSC

11


Fig. 3. Side by side image for stereoscopic HMD

4

Conclusion

In the virtual undersea world designed by this study, many fish and water plants are made in arbitrary virtual spaces, and fish movements are expressed by animation work of their basic motion patterns based on their characteristics. We have constructed a more natural virtual undersea system by building an undersea environment consisting of fish and water plants and describing more than 300 fish in the undersea world. Also, the system generates 3D‐stereoscopic images, which are transmitted to the wireless HMD so that users can recognize the space. The purpose of this system is to enhance reality by realizing interaction with fish based on users’ gestures. For this, this study proposed artificial fish, their cluster actions, their movements etc. based on related studies so far, especially, Reynolds’ and Tu, Gate, and Barke’s studies. In particular, we were able to reduce the complexity of time by adding Avoidance and Path Following in cluster behavior, and maintain an image generation speed at 30 fps in expressing 300 individual fish of 30 kinds and 80 water plant objects of 10 kinds. Also, by adding the Blend, Priority, and Flock modules, which operate each action, we were able to not only reduce the complexity of time in clusters but also express natural cluster behavior. Another merit of this paper, the contribution of this study, is that it can enhance users’ sense of experience and immersion with events of controlling virtual fish based on users’ gestures in the interaction between an aquarium and users. We propose for future research that more realistic virtual spaces should be realized by composing an undersea world based on the food chain in the marine ecosystem.

12



Acknowledgment. This work was supported by the IT R&D program of MCST/MKE/IITA. [2008-F-030-02, Development of Full 3D Reconstruction Technology for Broadcasting Communication Fusion]

References 1. Hyun-Cheol Lee, Eun-Seok Kim, Nak-Keun Joo, and Gi-Taek Hur. Development of real time virtual aquarium system. International Journal of Computer Science and Network Security, 6(7):58-63,( 2006). 2. Craig W. Reynolds, “Flocks, herds and schools: A distributed behavioral model”, SIGGRAPH 87 Proceedings of the 14th annual conference on Computer graphics and interactive techniques, 25-34, (1987). 3. Xiaoyuan Tu and Demetri Terzopoulos, “Artificial fishes: physics, locomotion, perception, behavior”, SIGGRAPH 94 Proceedings of the 21st annual conference on Computer graphics and interactive techniques, 43-50, (1994). 4. Peter Tarasewich and Patrick R. McMullen, “Swarm intelligence: power in numbers”, Communications of the ACM, 45(8) 62-67, (2002). 5. William R Sherman and Alan B Craig. Understanding virtual reality: Interface, application, and design. Morgan Kaufmann,12 (2002). 6. Tapio Takala, Lauri Savioja, and Tapio Lokki. Swimming in a virtual aquarium, http://www.academia.edu/2744573/ Swimming_in_a_Virtual_Aquarium.(2005) 7. Martin Usoh, Kevin Arthur, Mary C Whitton, Rui Bastos, Anthony Steed, Mel Slater, and Frederick P Brooks Jr. Walking > walking-in-place> flying, in virtual environments. In International Conference on Computer Graphics and Interactive Techniques, volume 1999, 359-364,( 1999).


13