Computer Animation and Virtual Worlds

We present a novel approach to the creation of varied animations by interpolating a small set of input images in a reduced pose-space. The technique includes an N-way mapping technique, an efficient, rigidity preserving nonlinear blending function, improved extrapolation and a novel scattered data interpolation technique to manage the reduced pose-space. Together these allow users to intuitively and interactively control temporal behaviours with simple gestures.

In this paper, we present a novel approach to creating and choreographing character animation using low-cost accelerometers based on the performer's movements. The goal of our research is to provide untrained users with an accelerometer-based interface for interactively choreographing complex human motions. It captures the movements of the performer with accelerometers, recognizes and interprets these movements based on a mocap database, and determines the motion data required to reproduce the desired movement of the performance.

In this paper, we propose a novel interactive framework for quadrangulating a set of models collectively with compatible connectivity. Furthermore, we demonstrate its application to 3D mesh morphing. In our approach, the user interactively sketches a skeleton within each model, and our method automatically computes compatible base domains for all models from these skeletons, on which the models are parameterized.

In this paper, we propose a framework to add gaze attention behaviors to crowd characters. It first consists in automatically detecting the points of interest which should be looked at by the characters. It then consists in a dedicated Inverse Kinematics solver to adapt the character motions in order for them to perform the gaze motions in a smooth and natural way.

Generating animation skeleton of a hand model with our method: (a) Original mode, (b) Primary 3D silhouette, (c) 3D medial axis of hand through constrained Delaunay triangulation, (d) Decomposition result, (e) Curve skeleton and key skeleton points, (f) Animation skeleton.

In this paper we introduce a real-time two step method for deforming geometric textures with dynamic effects. The major technical contributions include a shell deformation method which is aware of Poisson's ratio and a vibration simulation algorithm for geometric textures which support variant effects. The simulation can be both driven by an animation sequence of the base mesh or directly manipulating the geometric textures.

We propose a 3D stylized lighting method, focusing on several stylized effects including straight lighting, edge lighting, and detail lighting which are important features in hand-drawn cartoon animation. To accomplish these, we introduce a light coordinate system for smooth animation of light and shade. We also extend a toon mapping process for detailed feature lighting.

This paper presents a real-time, dynamic skin deformation technique driven by recorded EMG data and a displacement map created from an accurate 3D muscle model. This results in more realistic looking simulations that are very appropriate for games and virtual reality applications.

We present a new method for the realistic real-time simulation of rivers. Our solution includes a 2D fluid solver that simulates the flow of a river's surface, an efficient method for adaptively computing 3D flow information and an animated 3D procedural wave texture that is advected through the fluid via advection particles in order to mimic the highly detailed fluid surfaces that are characteristic of rivers.

This paper presents a new method which allows user to manipulate a two-dimensional shape in an intuitive and flexible way. Based on a hybrid shape representation, the method can produce the natural deformation results subject to the user s constraints. Meanwhile, the proposed approach provides user an intuitive and simple way to adjust the shape global and local stiffnesses. In addition, the method could be easily extended to manipulate curves and stick figures.

This paper presents the fireworks controller, a novel real-time shape-constrained fireworks animation system. We propose highly parallel point sampling and iterative clustering algorithms to support multi-level fireworks explosion. Several novel intuitive user interfaces are introduced to improve the usability of the system.

Systematic experimentation showed that user engagement with embodied agents is not only established through the traditional dimensions of aesthetics and realism but more so through the agent's action possibilities (affordances). Reversely, willingness to use the embodied agent is not only fueled by usability issues but equally so by the moral intentions designed into the agent (ethics). Results are compiled in the theoretical framework called Interactively Perceiving and Experiencing Fictional Characters (I-PEFiC).

This paper presents the angle-split shell textures (ASST) method for modelling and rendering fur with a wide variety of furstyles. This method classifies the shell textures into different types with different numbers of texture layers, by splitting the angle space of the viewing angles between fur orientation and view direction. Our system can render the fur with biological patterns, and utilize ASST to control the geometric variations of the furry shape.

A new method is presented to solve rigid body interaction in SPH. The new method is impulse-based and it gives more stable and plausible results than the conventional method in fluid-rigid and rigid-rigid interactions. Our SPH solver with this method may be regarded as a particle-based rigid body solver capable of two-way coupling to fluid and rigid body.

This paper proposes a simple and effective framework to deal with the problem of synthesizing interactive and competitive motions while reflecting the interactions.

In this paper we present a Chinese Sign Language Markup Language (CSLML), which integrates prosodic information into grammatical regulation to provide a more effective way in describing the features of Chinese Sign Language. Besides, CSLML is divided into two layers, function level and phonetic level, the mapping rules between them can be diversified to increase the flexibility and expressivity of synthesis.

This paper presents a novel approach for path-based smoke simulation that explores the theory of chemical kinetics in an aim to provide a useful animation tool. By describing intended smoke effects through chemical reaction equations and adjusting their parameters, it allows to easily create various interesting smoke effects that were often hard to get with previous techniques. We describe several examples of path-based smoke animations, generated with easily understandable reaction equations and control parameters.

The paper proposes a development of a computational cognition architecture for deriving behavioural models through three process models – reflexive, reactive and reflective, hence providing the infrastructure for virtual characters to build up a broad range of intellectual capabilities with processes that ranges from highly reactive to highly deliberative. The effect of the three process models is demonstrated in an implementation of a virtual tour application – •TOWN.

A new distance metric is proposed to estimate the 3D pose similarity in accordance with human perception. The distance metric is composed of geometric features selected from an extensive feature pool by Adaboost. Higher precision (judged by blind human perception) is achieved in the experiments, and the new distance metric gets best performance when adapted to specific motion types.

We present a static equilibrium model of a curved cantilever beam to bend branches or petioles, and we extend it to deform flexible laminae. Combined with a mass spring system-based dynamic model, we generate a pseudo-dynamics model and analyze the combination cases of natural frequency and damping ratio. Finally, we give some guidelines to determine the combination types of branches and leaves according to their structures and implement their motion in wind fields with different model parameters.

The pipeline of 3D face symmetrization. We present a novel method for enhancing the symmetry of a scanned 3D face automatically. We developed a new local 3D shape descriptor based on facial features and surface curvatures. Our shape descriptor can improve the accuracy when deforming a 3D face towards a symmetric configuration, because it provides accurate point pairing with respect to the plane of symmetry. In addition, we use point-based representation over all stages of symmetrization without generating a consistent triangle mesh or texture parameterization.

It is a fundamental but challenging problem to efficiently edit complex 3D objects. By embedding the input models into coarse tetrahedral control meshes, this paper develops a unified framework to discuss two useful editing operations: interactive deformation and deformation transfer. Experimental results show our algorithm is effective, easy to control, supports various shape representations, such as meshes, point clouds, models with multiple parts or non-manifold structures etc, and well transfers deformations between non-homeomorphous models.

A novel pressure correction scheme is proposed to be used along with the equations of state in SPH. With the pressure correction, the local pressure disturbance is transported to the neighboring area and more accurate and efficient fluid simulation is achieved.

We present a layered geometric approach to the skinning of character animation. On top of a global shape deformation, B-spline surface patches are attached to various body parts for local control. The patches are directly manipulated so as to generate example local shapes at some important poses. During character animation, the B-spline control points move continuously by interpolating a set of example patches kinematically, and at the same time they can also move dynamically by elastic simulation.

This paper presents a novel method to simulate skin sliding at real-time by remeshing the surface based on a parameter space resampling. In order to evaluate the surface parametrization, we use the force density method which solves for an energy minimizing form with a sparse linear system. Our method creates a realistic approximation of skin sliding in real-time, reducing texture distortions. In addition it is flexible, simple to use, and can be incorporated into any animation pipeline.

The proposed TFAN (Triangle Fan-based) codec supports real-time encoding/decoding of 3D meshes with arbitrary topologies (i.e. manifold or not, oriented or not), by specifying ten topological configurations (cf. Figure 1). TFAN outperforms existing techniques such as MPEG-4/3DMC or Touma & Gotsman, with decoding times lower by an order of magnitude at equivalent or even better levels of compression efficiency. Due to its high performances, the TFAN approach has been recently retained for integration within the MPEG-4/AFX standard.

A time-critical collision handling approach for deformable objects is presented. Detection, penetration depth estimation, and response are divided into atomic tasks. In case of an interruption, the algorithm basically resumes in the next time step. It is ensured that persistent collisions are handled in a subsequent simulation step. Approximate penalty forces are provided for colliding points whose collision handling is postponed to the next time step. The results are physically plausible and given time constraints can be guaranteed.

This paper presents a fast method that computes deformations with fracturing of an object using a hierarchical lattice. The present method employs adaptive shape matching with special care on stiffness control so that it yields deformation results similar to those using a uniform lattice, with lower computational cost. It also reduces the computational cost by merging object parts that are separated off by fracturing.

A novel method to automatically generate synchronized dance motion that is perceptually matched to a given musical piece is described. A matching process was performed between the musical and motion feature spaces considering the correspondence of the relative changes in both feature spaces and the correlations between musical and motion features. A user opinion study was designed and carried out to demonstrate the effectiveness of the proposed approach.

We propose a new method to concatenate two dynamic full-body motions such as punches, kicks, and flips by using the angular momentum as a cue. Through the observation of real humans, we have identified two patterns of angular momentum that make the transition of such motions efficient. Based on these observations, we propose a new method to concatenate two full-body motions in a natural manner.

Our shadowing approach is easy and intuitive, and renders shadows automatically by setting character's image sequence with alpha value as input. After rendering the shadows, animators can edit the rendered shadows for casting direction and shape globally, and then edit shadows locally by applying Gaussian Filter to blur the rendered shadows.

In Mixed Reality applications that use video see-through head-mounted display technology, blue-screening often identifies locations where virtual content should interact with the real environment. We propose a semi-automated, noise-tolerant algorithm capable of interactively keying video signals on a GPU. We show that our method, based on principle components analysis, produces alpha mattes comparable to commercial off-line solutions.

We present a new adaptive sampling model based on SPH framework for real-time fluid simulation by introducing a generalized distance field function in non-uniform particle system and building the splitting and merging rules for particles. To make the simulation more stable and visually continuous, we introduce a modified pressure state equation and a new surface tension model. Various fluid scenes including breaking wave and flood are simulated in real-time with realism.

Generating realistic behaviors for the non-player characters (also known as bots) is an important task for Military Operations on Urban Terrain (MOUT) simulations. In such situations, due to time constraints and incomplete information, humans rely mainly on experience rather than some structured analysis of the given situation to make decisions. To address this issue, we had incorporated expectations in SNAP, a time critical decision-making framework which aims to imitate human decision-making processes for MOUT simulations.

Despite robots' physical embodiment, the effect of communicative touch has been a mostly overlooked aspect of human-robot interaction. This video-based, 2 × 2 between-subject survey experiment (N = 119) found that the combination of touch and proactivity influenced whether people saw a robot as machine-like and dependable. Results show that communicative touch could be more appropriate for proactive agents than reactive agents. These results illustrate that careful consideration is necessary when incorporating social behaviours in agents' physical interaction design.