Search

This poster reports our first approach for automatic music arrangement based on the impression of input images. Given a digital image and keywords of objects shot in the image, the technique selects a rhythm pattern associated from the keywords and color distribution of the image. As a preprocessing, the technique first provides sample colors, images, and keywords to users, and then collects the feedback of selection of rhythm patterns associated from them. The technique then leads equations to calculate the association of rhythm pattern from arbitrary input images. Finally, the technique automatically calculates the association scores of all prepared rhythm patterns from the images, and provides music arranged applying the associated rhythm pattern. [ABSTRACT FROM AUTHOR]

When rendering immersive virtual environments, it is necessary to limit the geometric complexity of the scene in order to retain interactive frame rates. Both reduced frame rendering rates and reduced geometric complexity have been shown to have a negative impact on task performance. We present a perceptually adaptive rendering system which optimizes rendering performance in immersive virtual environments. Our system currently uses three discrete levels of detail for each model, which are created prior to run time. We use head position and orientation in the 3D environment to determine the appropriate geometric level of detail (LOD) of the objects in the scene. Based upon the thresholds used to trigger a change between two LODs, interactivity can be significantly improved without overtly distracting the user. [ABSTRACT FROM AUTHOR]

Animated virtual humans created by skilled artists play a remarkable role in today's public entertainment. However, ordinary users are still treated as audiences due to the lack of appropriate expertise, equipment, and computer skills. We developed a new method and a novel sketching interface, which enable anyone who can draw to "sketch-out" 3D virtual humans and animation. We devised a "Stick Figure→ Fleshing-out→Skin Mapping" graphical pipeline, which decomposes the complexity of figure drawing and considerably boosts the modelling and animation efficiency. We developed a gesture-based method for 3D pose reconstruction from 2D stick figure drawings. We investigated a "Creative Model-based Method", which performs a human perception process to transfer users' 2D freehand sketches into 3D human bodies of various body sizes, shapes and fat distributions. Our current system supports character animation in various forms including articulated figure animation, 3D mesh model animation, and 2D contour/NPR animation with personalised drawing styles. Moreover, this interface also supports sketch-based crowd animation and 2D storyboarding of 3D multiple character interactions. A preliminary user study was conducted to support the overall system design. Our system has been formally tested by various users on Tablet PC. After minimal training, even a beginner can create vivid virtual humans and animate them within minutes. [ABSTRACT FROM AUTHOR]

This paper describes a real-time system to synthesize a dancing character that is matched with beatboxing sounds. It is easier for humans to make a rhythmical sound with their mouth and vocal like a beatboxing than playing musical instruments. The system extracts rhythm features from dancing motion clips contained in a motion clip database at a preprocessing step. While performing beatboxing in real time, the system extracts rhythm features from those sounds and then finds appropriate a clip in the database whose rhythm features are matched to the sounds. [ABSTRACT FROM AUTHOR]

Texture synthesis aims to seamlessly extend a sample texture. This can be achieved by reproducing the texture's recognisable patterns along with minimising the occurrences of undesirable image characteristics such as disjoint image features. Presently, patch-based methods have consistently produced the most impressive results. These methods involve aligning and stitching texture patches together. However, image inconsistencies exist when image features are misaligned; they are either fused together or abruptly cut off. Our method minimises the occurrences of these anomalies by using image edges to align patches and redefine their boundaries to preserve texture features. This implementation requires little to no user-defined variables or iterations and can be practically scalable given some pre-computation. [ABSTRACT FROM AUTHOR]

For the generation of freeform 3D models, one of the most intuitive solution is to use sketch-based modeling environments. Unfortunately, since the user interface relies upon the analyse of sketches in order to determine which action is requested by the user, the possible amount of different operations can be limited. In this paper, we present a 3D sketching system based on multiple views. Each view is specialized on a component of the modeling process (like the skeleton, the profile, etc.), and is based on specific sketching interactions. With this approach, an user could improve its understanding of the modeling process and perform a larger range of modeling operations. [ABSTRACT FROM AUTHOR]

We developed a system that visualizes how design of interior space affects user's subjective states when they are communicating personal information on a computer. After measuring the subjective reassurance of users while they were communicating information, we used the results to plot and create personal information spaces. Then the subjective aspects of workspace design were visualized based on these personal information spaces. Designers can intuitively understand the subjective aspects of space design visualized based on a user's subjective states using our system. [ABSTRACT FROM AUTHOR]

This paper presents a new feature extraction method on the mesh model with some noise. Firstly, we compute Mesh saliency value of every vertex on the mesh model. Then we set it as the weighted factor and employ the weighted bi-quadratic Bézier surface to fit the vertex and its neighborhood by the least square method. So we obtain the smoothed principal curvature and principal direction of every vertex, which can be used for detecting and extracting the feature point. The advantage of our method is that we can achieve the better feature extraction result from the mesh model with some noise. [ABSTRACT FROM AUTHOR]

In this paper we introduce concepts in order to enhance the resolution of autostereoscopic displays for performing routine tasks. Our Sim2D3D technique presented earlier allows users to work with any 2D or 3D application in an autostereoscopic desktop environment. We have conducted a user study to investigate the impact of this technique on routine tasks performed using autostereoscopic displays. The problems identified in this user study led us to further improve the design of autostereoscopic user interfaces by increasing the effective resolution for context areas. [ABSTRACT FROM AUTHOR]

Multi-modal capabilities can be added to a simulation system in order to enhance data comprehension. We describe a system for adding sonification capabilities to a real-time computational fluid dynamics (CFD) simulator. Our system uses Max/MSP modules to add sonic properties to CFD solutions. The enhancements described in this paper allow users to locate sound sources in a 3-D environment using stereo auditory cues to identify data features. [ABSTRACT FROM AUTHOR]

This article gathers and consolidates the issues involved in uncertainty relating to reasoning and analyzes how uncertainty visualizations can support cognitive and meta-cognitive processes. Uncertainty in data is paralleled by uncertainty in reasoning processes, and while uncertainty in data is starting to get some of the visualization research attention it deserves, the uncertainty in the reasoning process is thus far often overlooked. While concurring with the importance of incorporating data uncertainty visualizations, we suggest also developing closely integrated visualizations that provide support for uncertainty in reasoning. [ABSTRACT FROM AUTHOR]

Generally, 2D computer graphics are suitable for depicting 2D structures. However, computer games generally depict 3D structures using 2D computer graphics. We are interested in such expression. As the overhead view has no 3D geometry data, a mechanism 3D geometry data from the overhead view is proposed. In addition, a tile-based technique for seamless conversion of a two-dimensional graphical expression into a three-dimensional geometric shape is proposed. In the proposed approach, an overhead 2D view defined by tiles on a canvas is converted into groups of 3D forms automatically and distorted according to an arbitrary viewing angle.The derivation and algorithm of the scheme are presented, and example applications are shown. [ABSTRACT FROM AUTHOR]

Data hiding techniques onto images provide tools for protecting copyright or sending secret messages, and they are currently utilized as a simple input device of a cell phone by detecting a data embedded in an image with an equipped digital camera. This paper presents a method of synthesizing texture images for embedding arbitrary data by utilizing the smart techniques of generating repetitive texture patterns through feature learning of a sample image. We extended the techniques so that a synthesized image can effectively conceal the embedded pattern, and the pattern can be robustly detected from a photographed image. We demonstrate the feasibility of our techniques using texture samples including an image scanned from real material. [ABSTRACT FROM AUTHOR]

This article describes a new interaction model for artistic shape generation. Instead of using a metaphor of the classical sculpting tools we propose a new approach based on a two level editing which both uses genetic programming methods: objects are created by means of subjective selection and generative evolution. We develop a new way to define these objects combining algebraic definition of implicit surfaces and CSG-like composition of these shapes. Along with these we propose a software tool that implements a simple way to manage this new approach in three dimensional objects design. [ABSTRACT FROM AUTHOR]

This paper introduces a semantic representation for virtual proto typing in interactive virtual construction applications. The representation reflects semantic information about dynamic constraints to define objects' modification and construction behavior as well as knowledge structures supporting multimodal interaction utilizing speech and gesture. It is conveniently defined using XML-based markup for virtual building parts. The semantic information is processed during runtime in two ways: Constraint graphs are mapped to a generalized data-flow network and scene-graph. Interaction knowledge is accessed and matched during multimodal analysis. [ABSTRACT FROM AUTHOR]

Automatic lighting design aims to provide users with semi-automated approaches, and easy-to-use tools, to configure lighting for 3D scenes. We present LightOpex the first fully automatic example-based local illumination lighting design system. Utilizing a wavelet-based lighting design framework, by which image quality is modeled using a cognitively inspired objective function, this approach to lighting design both: (1) allows the declarative specification of lighting; and (2) facilitates intuitive and natural specification of scene lighting. LightOpex enables users to select the desired lighting for a scene using exemplar 2D images and uses the spatial distribution of the luminance in these images as the target values of an optimization step. We demonstrate the utility of LightOpex through a range of examples, and conduct a preliminary investigation of the performance of a number of different of optimization schemes. [ABSTRACT FROM AUTHOR]

This paper presents the concept and an evaluation of a novel approach to support students to understand complex spatial relations and to learn unknown terms of a domain-specific terminology with coordinated textual descriptions and illustrations. Our approach transforms user interactions into queries to an information retrieval system. By selecting text segments or by adjusting the view to interesting domain objects, learners can request additional contextual information. Therefore, the system uses pre-computed multi-level representations of the content of explanatory text and of views on 3D models to suggest textual descriptions or views on 3D objects that might support the current learning task. Our experimental application is evaluated by a user study that analyzes (i) similarity measures that are used by the information retrieval system to coordinate the content of descriptive texts and computer-generated illustrations and (ii) the impact of the individual components of these measures. Our study revealed that the retrieved results match the preferences of the users. Furthermore, the statistical analysis suggests a rough value to cut-off retrieval results according to their relevancy. [ABSTRACT FROM AUTHOR]

Obesity and lack of physical fitness are increasingly common in adults as well as children and can negatively affect health. Regular physical activity, such as jogging or training in a fitness center, is recommended by physiologists to fight obesity and improve one's fitness, but usually requires considerable motivation. Recently, researchers as well as companies have proposed a few fitness games, i.e. videogames where users play by performing physical exercises, in which game elements (such as graphics and gameplay) are used to encourage people to exercise regularly. This paper proposes a fitness game system which aims at combining arcade-style game graphics, physiological sensors (e.g. heart rate monitor, 3D accelerometer), and an adaptation engine. The adaptation engine considers personal information provided by the user (e.g., age and gender), her current heart rate and movements, and information collected during previous game sessions to adjust the required intensity of physical exercises through context-aware and user-adaptive dynamic adaptations of graphics and gameplay. Besides describing the general system, the paper presents two implemented games and a preliminary user evaluation, which also led us to introduce in the system a 3D virtual human. [ABSTRACT FROM AUTHOR]

This research proposes 3D graphical agents in the role of virtual presenters with a new type of functionality - the capability to process and respond to visual attention of users communicated by their eye movements. Eye gaze is an excellent clue to users' attention, visual interest, and visual preference. Using state-of-the-art non-contact eye tracking technology, eye movements can be assessed in a unobtrusive way. By analyzing and interpreting eye behavior in real-time, our proposed system can adapt to the current (visual) interest state of the user, and thus provide a more personalized, context-aware, and ‘attentive' experience of the presentation. The system implements a virtual presentation room, where research content of our institute is presented by a team of two highly realistic 3D agents in a dynamic and interactive way. A small preliminary study was conducted to investigate users' gaze behavior with a non-interactive version of the system. A demo video based on our system was awarded as the best application of life-like agents at the GALA event in 2006. [ABSTRACT FROM AUTHOR]

Interactive control of 3D avatar motions has important applications in computer games and animations. Most of the current games directly map the inputs of a user into canned motions of the 3D character under control. However, an ideal user interface should be more like a human moving in the real life where appropriate motions are chosen according to the environment in the front. In this paper, we have developed an intelligent control interface that helps the user to look ahead a few steps and search a motion library organized in a motion graph for the most appropriate motion for execution. The search is conducted incrementally with the best effort in each frame according to a given available time budget. We have developed a best-first strategy to maintain the search tree in a way that the number of useful nodes is kept as large as possible. Experiments on several example scenes have been conducted to illustrate how this mechanism can enhance the usability of 3D avatar control from a third-person view. [ABSTRACT FROM AUTHOR]

We propose a novel Human Computer Interaction (HCI) paradigm for volume visualization in projection-based immersive virtual environments (VEs). This paradigm is intuitive, highly efficient and allows accurate control over the virtual objects. A fine control mode for direct manipulation is proposed to address the low accuracy problem of virtual object manipulation in VEs. An agent object interaction method is proposed to provide more flexibility in manipulating the volume objects. A two-handed scaling method is proposed to conveniently scale the volume object along one, two, or three axes. Finally, a ghost object paradigm is proposed to address the motion constraint problem for virtual objects. An implementation using a 3-state tracked glove setup as the input interface is discussed. How basic functionality for volume visualization can be transferred from the 2D WIMP (Window, Icon, Menu, and Pointer) interface to a 3D VR interface is also systematically discussed. [ABSTRACT FROM AUTHOR]

In this paper, we present a fast and intuitive interface for sketching out 3D virtual humans and animation. The user draws stick figure key frames first and chooses one for "fleshing-out" with freehand body contours. The system automatically constructs a plausible 3D skin surface from the rendered figure, and maps it onto the posed stick figures to produce the 3D character animation. A "creative model-based method" is developed, which performs a human perception process to generate 3D human bodies of various body sizes, shapes and fat distributions. In this approach, an anatomical 3D generic model has been created with three distinct layers: skeleton, fat tissue, and skin. It can be transformed sequentially through rigid morphing, fatness morphing, and surface fitting to match the original 2D sketch. An auto-beautification function is also offered to regularise the 3D asymmetrical bodies from users' imperfect figure sketches. Our current system delivers character animation in various forms, including articulated figure animation, 3D mesh model animation, 2D contour figure animation, and even 2D NPR animation with personalised drawing styles. The system has been formally tested by various users on Tablet PC. After minimal training, even a beginner can create vivid virtual humans and animate them within minutes. [ABSTRACT FROM AUTHOR]

We introduce in this paper an interface for modeling botanical trees that we term a sketch-and-spray interface. In using the interface, a user would sketch out various types of strokes to construct the main structures of a tree. A user could use a branch stroke to draw a single branch, a leaf stroke to draw a single leaf or a stem of leaves, or a copy stroke to draw a whole branch together with all its sub-branches and leaves. A user could also use various editing strokes to modify the tree structure. Finally, a user could spray particles upon it to ‘grow' the overall shape of the tree. We illustrate the utility of our interface using a number of examples throughout this paper. [ABSTRACT FROM AUTHOR]

NPR Lenses is an interactive technique for producing expressive non-photorealistic renderings. It provides an intuitive visual interaction tool for illustrators, allowing them to seamlessly apply a large variety of emphasis techniques. Advantages of 3D scene manipulation are combined with the capabilities of viewer-aligned lenses by inserting 2D lens controls into the 3D rendering pipeline. By using lenses that are spatially explicit and constrained we enable object-independent adjustments in 3D. Regions of emphasis can be easily created by adjusting lens parameters and any combination of line style, type, shape, and position of computer-generated lines. NPR Lenses support the creation of effects similar to those found in many existing hand-drawn illustrations and allow creative freedom through interactive control at the same time. [ABSTRACT FROM AUTHOR]

This article proposes a sketch-based interface for modeling muscle fiber orientation of a 3D virtual heart model. Our target was electrophysiological simulation of the heart and fiber orientation is one of the key elements to obtaining reliable simulation results. We designed the interface and algorithm based on the observation that fiber orientation is always parallel to the surface of the heart. The user specifies the fiber orientation by drawing a freeform stroke on the object surface. The system first builds a vector field on the surface by applying Laplacian smoothing to the mesh vertices and then builds a volumetric vector field by applying Laplacian smoothing to the voxels. We demonstrate the usefulness of the proposed method through a user study with a cardiologist. [ABSTRACT FROM AUTHOR]

Application softwares usually support a single set of user interaction tools that are supposed to be optimal. It is satisfactory if the user interface (UI) is really optimal for everyone. However, since the user's intent and the preference vary one by one, such fixed toolkits may cause inefficiency of the UI. [ABSTRACT FROM AUTHOR]

While, based on shading, shadowing and outline etc, human beings generally perceive a shape, we explore an new representation for perceiving shape. The Invisible Shape is an interactive media art software which has a concept of a new expression for perceiving 3D shapes. Although our work does not draw a shape and the apparent size of the character does not change by parallel projection, the user may be able to perceive the shape by tracking the walking characters on it. Like subjective contour, we expect that the user perceives the surface of invisible shape by motion of the objects related to the shape. [ABSTRACT FROM AUTHOR]

Many computer games are based on "overhead" two-dimensional (2D) computer graphics which are suitable for displaying planar figure. However, in such graphical engines, three-dimensional (3D) structures are also presented in 2D. In the overhead view, the width of a shape is displayed by a horizontal line, and the height and depth are expressed as a vertical line. Such an expression in not realistic, but is intuitively recognizable, particularly in computer games. In this system, 3D geometry data recognizable by the user is generated from tiles placed on a canvas. The geometric shape created by the system can be viewed from an arbitrary angle by a gradual warping of the geometry, without the user being aware of the conversion to the 3D expression. [ABSTRACT FROM AUTHOR]