Computer Animation Research Articles

Dancing robots: aesthetic engagement is shaped by stimulus and knowledge cues to human animacy

IntroductionArtificial intelligence (AI) and robots are increasingly shaping the aesthetic preferences of art consumers, influencing how they perceive and engage with artistic works. This development raises various questions: do cues to the humanness of the origin of an artwork or artist influence our aesthetic preferences?.MethodsAcross two experiments, we investigated how the perception and appreciation of dance is influenced by cues to human animacy. We manipulated Agent Form (human-like or robot-like dancer), Belief about Movement Source (human motion capture or computer animation), Source of Choreography (human- or computer-generated), and Belief about Choreography Source (believed to be human- or computer-generated).ResultsResults pointed toward agent congruence: In Experiment 1, robot agents were preferred when the movement source was believed to be computer animation. In Experiment 2, robot agents were preferred when the choreography was believed to be computer-generated, while choreographies believed to be human-generated were generally preferred. Participants could not accurately identify the actual source of choreography. These results persisted beyond the effects of age, dance expertise, technological expertise, attitudes toward AI, and perceived familiarity, complexity, evocativeness, technical competence, or reproducibility of the dance. Dance expertise, technological expertise, and attitudes toward AI independently impacted aesthetic judgments.DiscussionThese findings provide insights into the design of robotic dance, highlighting features of dance choreography and audience characteristics that influence aesthetic engagement. To enhance AI-driven creative productions, shaping perceptions will be crucial for better audience reception and engagement.

A novel heterogeneous deformable surface model based on elasticity

The thin membranes and shells in nature are heterogeneous. They are widely used in surgical simulation, biological techniques, and computer animation. The corresponding surface deformable models can implement dynamic simulations of thin membranes and shells in nature, while most surface deformable models are isotropic and cannot represent thin membranes and shells in nature accurately. Therefore, we propose a novel physically-based heterogeneous deformable surface model. By utilizing the same B-spline basis functions or the parameter space of surfaces' geometric representations, we implement material modeling and propose the representations of surfaces with material variations with composite or continuous material functions. Then, we propose a novel physically-based elastic deformable surface model that constructs infinitesimal elements in the parameter space and employs elasticity to analyze their deformation. The corresponding elastic potential energy function is only related to surfaces' continuous representations, and our model avoids the computation error caused by meshes' quality and large rotation of points' frames. We employ isogeometric analysis to solve the dynamic equations derived from our surface model. To demonstrate the validity and reality of our model, several comparison experiments are designed. The corresponding results are in line with expectations and consistent with physical laws.

Cutting-edge techniques in 3D modeling and animation: Leveraging mathematical models and advanced software tools

Abstract. This paper explores the advancements in computer animation technology, focusing on 3D modeling and animation. It delves into the core aspects of character design, environment modeling, and animation rigging, emphasizing the mathematical models and algorithms that enhance realism and efficiency. Key techniques discussed include Bezier curves and spline interpolation for anatomy and proportions, Perlin noise and fractal algorithms for texturing, and Inverse Kinematics (IK) and Forward Kinematics (FK) for rigging. Additionally, the paper examines procedural generation and fluid dynamics simulations for environment modeling and the integration of motion capture data. The use of software tools like Blender, Autodesk Maya, and Houdini is highlighted. This study aims to provide a comprehensive understanding of the current state of 3D animation technology and its future directions.

Robust and Artefact‐Free Deformable Contact with Smooth Surface Representations

AbstractModeling contact between deformable solids is a fundamental problem in computer animation, mechanical design, and robotics. Existing methods based on C0‐discretizations—piece‐wise linear or polynomial surfaces—suffer from discontinuities and irregularities in tangential contact forces, which can significantly affect simulation outcomes and even prevent convergence. In this work, we show that these limitations can be overcome with a smooth surface representation based on Implicit Moving Least Squares (IMLS). In particular, we propose a self collision detection scheme tailored to IMLS surfaces that enables robust and efficient handling of challenging self contacts. Through a series of test cases, we show that our approach offers advantages over existing methods in terms of accuracy and robustness for both forward and inverse problems.

Diffusion‐based Human Motion Style Transfer with Semantic Guidance

Abstract3D Human motion style transfer is a fundamental problem in computer graphic and animation processing. Existing AdaIN‐based methods necessitate datasets with balanced style distribution and content/style labels to train the clustered latent space. However, we may encounter a single unseen style example in practical scenarios, but not in sufficient quantity to constitute a style cluster for AdaIN‐based methods. Therefore, in this paper, we propose a novel two‐stage framework for few‐shot style transfer learning based on the diffusion model. Specifically, in the first stage, we pre‐train a diffusion‐based text‐to‐motion model as a generative prior so that it can cope with various content motion inputs. In the second stage, based on the single style example, we fine‐tune the pre‐trained diffusion model in a few‐shot manner to make it capable of style transfer. The key idea is regarding the reverse process of diffusion as a motion‐style translation process since the motion styles can be viewed as special motion variations. During the fine‐tuning for style transfer, a simple yet effective semantic‐guided style transfer loss coordinated with style example reconstruction loss is introduced to supervise the style transfer in CLIP semantic space. The qualitative and quantitative evaluations demonstrate that our method can achieve state‐of‐the‐art performance and has practical applications. The source code is available at https://github.com/hlcdyy/diffusion-based-motion-style-transfer.

GAP FILLING ALGORITHM FOR MOTION CAPTURE DATA TO CREATE REALISTIC VEHICLE ANIMATION

The dynamic development of the entertainment market entails the need to develop new methods enabling the application of current scientific achievements. Motion capture is one of the cutting-edge technologies that plays a key role in movement and trajectory computer mapping. The use of optical systems allows one to obtain highly precise motion data that is often applied in computer animations. This study aimed to define the research methodology proposed to analyze the movement of remotely controlled cars utilizing developed gap filling algorithm, a part of post-processing, for creating realistic vehicle animation. On a specially prepared model, six various types of movements were recorded, such as: driving straight line forward, driving straight line backwards, driving on a curve to the left, driving on a curve to the right and driving around a roundabout on both sides. These movements were recorded using a VICON passive motion capture system. As a result, three-dimensional models of vehicles were created that were further post-processed, mainly by filling in the gaps in the trajectories. The case study highlighted problems such as missing points at the beginning and end of the recordings. Therefore, algorithm was developed to solve the above-mentioned problem and allowed for obtaining an accurate movement trajectory throughout the entire route. Realistic animations were created from the prepared data. The preliminary studies allowed one for the verification of the research method and implemented algorithm for obtaining animations reflecting accurate movements.

Early Image Processing and Computer Animation in Colorado: Pat Lehman’s Video Vitae in Context

Abstract In 1975, Pat Lehman screened her experimental video, Video Vitae at the New York Women’s Video Festival. Lehman worked in Colorado, where she also started the first video program in the state’s college system. This paper examines Video Vitae as an example of image-processed video and early computer animation in the context of second-wave feminism. The author charts Lehman’s influences and access to the Scanimate computer used to produce Video Vitae, arguing that the significance of Lehman’s feminist narrative and the technology were deeply intertwined. Such explicit feminist narratives were extremely rare among artists engaged in early image-processing and animation.

The use of animation tools in the media space of library for the formation of undergraduates' teamwork skills

The problem and the aim of the study. The modernization of library education involves the use of technologies that provide simultaneous work with sound, statistical images, and texts in an interactive mode. The ability to use information tools, in particular computer animation, in working on projects and teamwork skills are indemand soft skills, the maturity of which largely determines the professional self-realization of librarianship specialists. The purpose of the study is to identify the potential of using animation tools in the library's media space to develop the teamwork skills of master’s students. Research methods. The media space of the library is considered as a set of resources that allow professional collaboration of cultural institutions in a new format, attracting potential users to visit libraries, as well as informing them about their services. The Renderforest platform is used to develop didactic animation. The experiment involved 64 undergraduates of specialty 04.51.01 Cultural studies, profile – "Socio-cultural design", Vyatka State University (Russian Federation). The authors use V. Stefanson's methodology to assess the level of teamwork skills of master's students. For statistical processing, – Pearson's χ2 (chi-square) test. A group of experts has been involved in the evaluation of media projects for “Library Night” event. Fisher's method was used to establish statistically significant differences. Results. The students of the experimental group used animation tools to develop projects and library and information services according to the following algorithm: studying a digital service, defining the concept of an action/event for “Library Night”, writing a script, implementing a project, evaluating the potential for the development of the library's media space. Statistically significant differences in the qualitative changes that occurred in the system were revealed: according to the levels of formation of teamwork skills according to the criteria of dependence/non-dependence on the opinion of the group (χ2 = 6.427; p < 0.05); sociability/ nonsociability (χ2 = 7.879; p < 0.05); acceptance of "struggle" / "struggle" / avoidance of "struggle" (χ2 = 6.081; p < 0.05); and by the quality of training – φcrit=1.64<φemp=2.804. Conclusion. The possibilities of using animation tools in the media space of the library for the development of teamwork skills of master’s students are formulated: new formats of collaboration and project activities, innovative work experience, etc. However, there are also factors complicating the process: absence of clear leadership; constrains of communication; weak motivation; low social status of the profession.

Evaluation of The Flipped Classroom Model on Students' Academic Achievement and Retention in MAN 1 Ambon

This study evaluates the impact of computer simulation and animation instructional packages in natural sciences on high school student's academic achievement and retention in a flipped classroom setting. Using a quasi-experimental pre-test and post-test design, 30 students used the computer simulation package, and 29 students used the computer animation package. Data were collected via academic achievement and retention tests and analyzed using descriptive statistics and independent sample t-tests. The results showed significant improvements in academic achievement for both groups. The average post-test score increased from 28.52 to 70.35 in the computer simulation group and from 25.74 to 64.23 in the computer animation group. The t-test indicated no significant difference in academic achievement between the two groups (t = 1.98, p = 0.79). Retention scores showed a slight decrease from post-test to retention test but remained high, with average retention scores of 68.74 for the simulation group and 61.15 for the animation group. The t-test also revealed no significant difference in retention between the groups (t = 1.12, p = 0.66). The implications of this study suggest that using educational technology, such as computer simulations and animations, in a flipped classroom setting can significantly enhance students' academic achievement and knowledge retention. It underscores the importance of integrating technology into the educational curriculum to create a more inclusive, interactive, and effective learning environment. Additionally, these findings allow educators to choose the teaching methods that best suit their instructional needs and contexts, whether in resource-limited situations or when teaching complex concepts.

3D Character Animation and Asset Generation Using Deep Learning

Besides video content, a significant part of entertainment is represented by computer games and animations such as cartoons. Creating such entertainment is based on two fundamental steps: asset generation and character animation. The main problem stems from its repetitive nature and the needed amounts of concentration and skill. The latest advances in deep learning and generative techniques have provided a set of powerful tools which can be used to alleviate these problems by facilitating the tasks of artists and engineers and providing a better workflow. In this work we explore practical solutions for facilitating and hastening the creative process: character animation and asset generation. In character animation, the task is to either move the joints of a subject manually or to correct the noisy data coming out of motion capture. The main difficulties of these tasks are their repetitive nature and the needed amounts of concentration and skill. For the animation case, we propose two decoder-only transformer based solutions, inspired by the current success of GPT. The first, AnimGPT, targets the original animation workflow by predicting the next pose of an animation based on a set of previous poses, while the second, DenoiseAnimGPT, tackles the motion capture case by predicting the clean current pose based on all previous poses and the current noisy pose. Both models obtained good performances on the CMU motion dataset, with the generated results being imperceptible to the untrained human eye. Quantitative evaluation was performed using mean absolute error between the ground truth motion vectors and the predicted motion vector. For both networks AnimGPT and DenoiseAnimGPT errors were 0.345, respectively 0.2513 (for 50 frames) that indicates better performances compared with other solutions. For asset generation, diffusion models were used. Using image generation and outpainting, we created a method that generates good backgrounds by combining the idea of text conditioned generation and text conditioned image editing. A time coherent algorithm that creates animated effects for characters was obtained.

Synthesizing realistic sand assemblies with denoising diffusion in latent space

AbstractThe shapes and morphological features of grains in sand assemblies have far‐reaching implications in many engineering applications, such as geotechnical engineering, computer animations, petroleum engineering, and concentrated solar power. Yet, our understanding of the influence of grain geometries on macroscopic response is often only qualitative, due to the limited availability of high‐quality 3D grain geometry data. In this paper, we introduce a denoising diffusion algorithm that uses a set of point clouds collected from the surface of individual sand grains to generate grains in the latent space. By employing a point cloud autoencoder, the three‐dimensional point cloud structures of sand grains are first encoded into a lower‐dimensional latent space. A generative denoising diffusion probabilistic model is trained to produce synthetic sand that maximizes the log‐likelihood of the generated samples belonging to the original data distribution measured by a Kullback‐Leibler divergence. Numerical experiments suggest that the proposed method is capable of generating realistic grains with morphology, shapes and sizes consistent with the training data inferred from an F50 sand database. We then use a rigid contact dynamic simulator to pour the synthetic sand in a confined volume to form granular assemblies in a static equilibrium state with targeted distribution properties. To ensure third‐party validation, 50,000 synthetic sand grains and the 1542 real synchrotron microcomputed tomography (SMT) scans of the F50 sand, as well as the granular assemblies composed of synthetic sand grains are made available in an open‐source repository.

Electron Nuclear Dynamics of H+ + C2H2 at ELab = 30, 200, and 450 eV.

We present a complete simplest-level electron nuclear dynamics (SLEND) investigation of H+ + C2H2 at collision energies ELab = 30, 200, and 450 eV. This reaction is relevant in astrophysics and provides a computationally feasible prototype for proton cancer therapy reactions. SLEND is a time-dependent, variational, direct, and nonadiabatic method that adopts a classical-mechanics description for the nuclei and a Thouless single-determinantal wave function for the electrons. We perform this study with our code PACE, which incorporates the One Electron Direct/Electron Repulsion Direct (OED/ERD) atomic integrals package developed by the Bartlett group. Current SLEND simulations with the 6-31G** basis set involves 2,646 trajectory calculations from 9 nonredundant, symmetry-inequivalent projectile-target orientations. For H+ + C2H2 at ELab = 30 eV, SLEND/6-31G** simulations predict one simple scattering process, and three reactive ones: C2H2 hydrogen substitution, C2H2 fragmentation into two CH moieties, and C2H2 fragmentation into CHC and H moieties, respectively. We reveal and analyze the mechanisms of these processes through computer animations; this valuable chemical information is inaccessible by experiments. The SLEND/6-31G** scattering angle functions exhibit primary and secondary rainbow scattering features that vary with the projectile-target orientations and collision energies. SLEND/6-31G** predicts 1-electron-transfer (1-ET) integral cross sections at ELab = 30, 200, and 450 eV in good agreement with their experimental counterparts. SLEND/6-31-G** predicts 1-ET differential cross sections (DCSs) at ELab = 30 eV that agree well with their experimental counterparts over all the measured scattering angles. In addition, SLEND/6-31G** predicts 0-ET DCSs at ELab = 30 eV that agree well with their experimental counterparts at low scattering angles, but less satisfactorily at higher ones. Remarkably, both the 0- and 1-ET DCSs from SLEND/6-31G** exhibit distinct primary rainbow scattering signatures in excellent agreement with their experimentally inferred counterparts. Furthermore, both SLEND/6-31G** and the experiment indicate that the primary rainbow scattering angles from the 0- and 1-ET DCSs are identical (an unusual fact in proton-molecule collisions). Through these rainbow scattering predictions, SLEND has also validated a procedure to extract primary rainbow angles from structureless DCSs. We analyze the obtained theoretical results in comparison with available experimental data and discuss forthcoming developments in the SLEND method.

MPMNet: A Data-Driven MPM Framework for Dynamic Fluid-Solid Interaction.

High-accuracy, high-efficiency physics-based fluid-solid interaction is essential for reality modeling and computer animation in online games or real-time Virtual Reality (VR) systems. However, the large-scale simulation of incompressible fluid and its interaction with the surrounding solid environment is either time-consuming or suffering from the reduced time/space resolution due to the complicated iterative nature pertinent to numerical computations of involved Partial Differential Equations (PDEs). In recent years, we have witnessed significant growth in exploring a different, alternative data-driven approach to addressing some of the existing technical challenges in conventional model-centric graphics and animation methods. This article showcases some of our exploratory efforts in this direction. One technical concern of our research is to address the central key challenge of how to best construct the numerical solver effectively and how to best integrate spatiotemporal/dimensional neural networks with the available MPM's pressure solvers. In particular, we devise the MPMNet, a hybrid data-driven framework supporting the popular and powerful MPM, to combine the comprehensive properties of MPM in numerically handling physical behaviors ranging from fluid to deformable solids and the high efficiency of data-driven models. At the architectural level, our MPMNet comprises three primary components: A data processing module to describe the physical properties by way of the input fields; A deep neural network group to learn the spatiotemporal features; And an iterative refinement process to continue to reduce possible numerical errors. The goal of these special technical developments is to aim at involved numerical acceleration while preserving physical accuracy, realizing efficient and accurate fluid-solid interactions in a data-driven fashion. The extensive experimental results verify that our MPMNet can tremendously speed up the computation compared with the popular numerical methods as the complexity of interaction scenes increases while better retaining the numerical accuracy.

Stylizing Ribbons: Computing Surface Contours With Temporally Coherent Orientations.

Line work is a core element for the stylization of computer animations used by recent shows. However, existing stylization techniques are limited to edge treatments based on brush strokes or textures applied solely on top of curves. In this work, we propose new stylization effects by offering artists direct control over the inside and outside of surface contours. To this end, we introduce a method that creates ribbons, geometry strips of possibly varying width, that extrude from each side of the surface contour with temporally coherent orientations. Our contributions include the generation of spatially and temporally consistent normal orientations along visible contours and a trimming routine that converts arrangements of offset curves into ribbons free of intersections. We demonstrate the expressiveness and versatility of stylized ribbons by applying various effects on both character and shadow edges from animation sequences.

Effect of Computer Animation Instructional Package on Students’ Achievement in Hybridization in Chemistry

The study aimed to ascertain how learners' performance in the chemical hybridization concept was affected by the Computer Animation Instructional Package (CAI). A pretest-posttest quasi-experimental research design was used in the study. There were 120 seniors secondary school students in the sample. The control group received instruction via traditional lectures, while the experimental group was treated using computer animation. The Hybridization Achievement Test (HAT) and Students' Spatial Ability Test (SSAT) were tested and used for data collection. The tests had reliability coefficients of 0.71 and 0.77, respectively using Kuder Richardson 20. Using estimated marginal means and Analysis of Covariance (ANCOVA) at a 0.05 significance level, seven research hypotheses were tested. Results showed that CAI highly impacted students' performance in the chemical concept of hybridization. A noteworthy interaction effect between gender and spatial ability was also found, favouring females with high spatial ability over males with low spatial ability regarding learners' performance. The research results indicate that CAI enhanced students' performance on the hybridization concept in chemistry. The results also showed the gender-friendly and successful instructional characteristics of CAI.

Geometric Modelling of a Family of 4-Point Ternary Approximating Subdivision Scheme U_φ with Visual Performance

Making signals better than noise in communication has always been challenging for scientists. Researchers have been working on it in different ways. The computer-aided geometric design is a new research field emerging from the collaboration of computer algorithms and mathematical logic towards curve designing, in which the subdivision schemes used have a key position due to their flexible and smooth behaviour. Using parameters in these schemes allows for increased control over designing. A parameterized framework for generating a wide range of subdivision surfaces with tunable degrees of shape control is presented in the family of schemes. The properties of the proposed family make it suitable for use in isogeometric analysis, computer animation, and geometric modelling. The purpose of this paper is to construct and analyze a family of 4-point ternary subdivision schemes to smooth the curves based on the Laurant polynomial. This family is generated by tuning the weight parameter. The scheme is analysed for its different properties. The scheme has continuity. Visual performance of the subdivision scheme is also provided as an application of this proposed study.

The effect of 7E learning cycle enriched with computer animations on students conceptual understanding and overcoming misconceptions

The effect of 7E learning cycle enriched with computer animations on students conceptual understanding and overcoming misconceptions

Animation in digital media

The article is devoted to the creation of animated works in the digital environment. Modern technologies develop new approaches to the formation of communication between the viewer and the work. Analysis of the new creation technologies for computer animation reveals that animation performed on game engines with the use of motion capture and computer graphics transforms the production process, with characters “animated” via digitizing of the actor’s movement. Actor's facial expressions and movements conveyed to the digital character depend on the actor's skill in portraying their character's personality. This develops limitations of creating animated plastic movement, exaggeration, hyperbolization, and conventionality, thus making animators look for new approaches to visual imagery, as well as opportunities to improve the means of artistic expression for the modern computer animation.

A language‐directed virtual human motion generation approach based on musculoskeletal models

AbstractThe development of the systems capable of synthesizing natural and life‐like motions for virtual characters has long been a central focus in computer animation. It needs to generate high‐quality motions for characters and provide users with a convenient and flexible interface for guiding character motions. In this work, we propose a language‐directed virtual human motion generation approach based on musculoskeletal models to achieve interactive and higher‐fidelity virtual human motion, which lays the foundation for the development of language‐directed controllers in physics‐based character animation. First, we construct a simplified model of musculoskeletal dynamics for the virtual character. Subsequently, we propose a hierarchical control framework consisting of a trajectory tracking layer and a muscle control layer, obtaining the optimal control policy for imitating the reference motions through the training. We design a multi‐policy aggregation controller based on large language models, which selects the motion policy with the highest similarity to user text commands from the action‐caption data pool, facilitating natural language‐based control of virtual character motions. Experimental results demonstrate that the proposed approach not only generates high‐quality motions highly resembling reference motions but also enables users to effectively guide virtual characters to perform various motions via natural language instructions.

Graph‐based control framework for motion propagation and pattern preservation in swarm flight simulations

AbstractSimulation of swarm motion is a crucial research area in computer graphics and animation, and is widely used in a variety of applications such as biological behavior research, robotic swarm control, and the entertainment industry. In this paper, we address the challenges of preserving structural relations between the individuals in swarm flight simulations by proposing an innovative motion control framework that utilizes a graph‐based hierarchy to illustrate patterns within a swarm and allows the swarm to perform flight motions along externally specified paths. In addition, this study designs motion propagation strategies with different focuses for varied application scenarios, analyzes the effects of information transfer latencies on pattern preservation under these strategies, and optimizes the control algorithms at the mathematical level. This study not only establishes a complete set of control methods for group flight simulations, but also has excellent scalability, which can be combined with other techniques in this field to provide new solutions for group behavior simulations.