Implicit distance-field representations, such as signed and unsigned distance fields (SDFs and UDFs), have become fundamental tools for learning and modeling 3D geometry. However, when applied to segmented closed surfaces—closed solids decomposed into many labeled regions with complex internal interfaces, as in brittle fracture fragments or multi-material parts—standard UDF-based formulations become increasingly difficult to optimize when many small fragments are present, often requiring substantially longer training and making numerical convergence a poor indicator of whether small-fragment geometry has truly converged. We introduce Far-From-Boundary Fields (FFBFs), a task-specific boundary-aware scalar field representation for segmented solids. FFBF is obtained by per-fragment reparameterization of a segmented UDF formulation, while preserving the zero level sets and fragment labeling of the underlying solid. This yields a more balanced and stable implicit field in which fragment-wise boundaries are easier to capture, and treats segmentation as a generative distance-field-based representation rather than a separate label prediction problem. Across brittle fracture benchmarks, we show that FFBF consistently improves geometric accuracy, small-fragment recovery, and internal-boundary consistency compared to direct UDF baselines and common variants. These results indicate that FFBF provides a useful reparameterization for learning-based modeling of segmented volumetric structures, including applications to data-driven brittle fracture prediction.

Abstract Tall cell grids have been proposed as an efficient approach to accelerate large‐scale liquid simulation. In this framework, regions near the liquid surface are discretized with regular grids, while regions farther away are represented by elongated rectangular cells. The regular grid region close to the surface is referred to as the optical layer. In previous work, the thickness of this optical layer was uniformly fixed across the entire liquid domain. In this paper, we propose a novel tall cell grid structure in which the thickness of the optical layer is dynamically adjusted according to the motion of the liquid. This adaptive strategy reduces the number of grid cells required in the projection step without compromising visual quality, thereby accelerating the overall simulation. Furthermore, we introduce a two‐way coupling scheme between rigid bodies and liquids in regions where the optical layer remains thin. Our algorithm is simple and can be easily integrated into existing tall cell grid frameworks.

ABSTRACT Realistic fluid simulations in interactive applications require both high visual fidelity and the ability to incorporate new obstacles at run time. In practice, high‐quality Eulerian fluid data are often precomputed offline, but inserting obstacles into such baked fields typically requires solving a global pressure Poisson equation, negating the efficiency gains of precomputation. We present a framework for obstacle‐aware fluid control based on vector potential formulation, which maintains divergence‐free velocity fields by construction. Our method introduces a projection‐free boundary handling technique inspired by Curl‐Noise, where a one‐sided potential field decomposition enforces approximate free‐slip conditions around arbitrary obstacles without global re‐solves. To restore turbulent details suppressed by numerical dissipation, we further incorporate a Vortex Primitive Method (VPM) that injects physically‐informed vortex particles at separation points identified through surface curvature criteria and a Bernoulli‐inspired pressure heuristic. The VPM operates in vector potential space via the Biot‐Savart law, preserving the divergence‐free property of the reconstructed velocity field. Experiments on three scenarios with comprehensive quantitative metrics demonstrate that our per‐frame editing cost is two orders of magnitude lower than a standard pressure Poisson solve, enabling near real‐time one‐way fluid‐solid interaction with near‐exact mass conservation.

ABSTRACT Real‐time character animation requires generating natural motions while preserving diverse walking styles under user control. Phase‐based representations are commonly employed in motion generation frameworks to control periodic motions such as walking; however, existing approaches face a trade‐off between preserving fine‐grained local periodic details and maintaining coherent whole‐body motion. Methods focusing on local periodic features often insufficiently represent other joints, while global phase representations tend to smooth out stylistic details, leading to style homogenization. This paper proposes an end‐effector‐aware phase manifold learning framework that balances local stylistic features and global motion consistency. The proposed method employs a two‐stage training strategy that first learns local periodic characteristics of end‐effectors and then integrates full‐body periodicity while fixing the learned local representations. In addition, we introduce an angular velocity‐based phase representation, which more directly captures the rotational characteristics of walking motions than linear velocity. Experimental results demonstrate improved style preservation for gesture‐dominant motions while generating stable whole‐body walking motions.

Abstract In the field of brittle fracture animation, generating realistic destruction animations using physics‐based simulation methods is computationally expensive. While techniques based on Voronoi diagrams or pre‐fractured patterns are effective for real‐time applications, they fail to incorporate collision conditions when determining fractured shapes during runtime. This paper introduces a novel learning‐based approach for predicting fractured shapes based on collision dynamics at runtime. Our approach seamlessly integrates realistic brittle fracture animations with rigid body simulations, utilising boundary element method (BEM) brittle fracture simulations to generate training data. To integrate collision scenarios and fractured shapes into a deep learning framework, we introduce generative geometric segmentation, distinct from both instance and semantic segmentation, to represent 3D fragment shapes. We propose an eight‐dimensional latent code to address the challenge of optimising multiple discrete fracture pattern targets that share similar continuous collision latent codes. This code will follow a discrete normal distribution corresponding to a specific fracture pattern within our latent impulse representation design. This adaptation enables the prediction of fractured shapes using neural discrete representation learning. Our experimental results show that our approach generates considerably more detailed brittle fractures than existing techniques, while the computational time is typically reduced compared to traditional simulation methods at comparable resolutions.

Abstract We present a unified, primitive‐first framework with DCD for collision response in physics‐based simulations. Previous methods do not provide sufficient solutions on a framework that resolves edge‐triangle and edge‐edge collisions when handling self‐collisions and inter‐object collisions in a unified manner. We define a scalar function and its gradient, representing the distance between two triangles and the movement direction for collision response, respectively. The resulting method offers an effective solution for collisions with minor computational overhead and robustness for any type of deformable object, such as solids or cloth. The algorithm is conceptually simple and easy to implement. When using PBD/XPBD, it is straightforward to incorporate our method into a collision constraint.

We present a novel solution for mesh-based deformation simulation from a spectral perspective. Unlike existing approaches that demand separate training for each garment or body type and often struggle to produce rich folds and lifelike dynamics, our method achieves the quality of physics-based simulations while maintaining superior efficiency within a unified model. The key to achieve this lies in the development of a spectrum-enhanced deformation network, a result of in-depth theoretical analysis bridging neural networks and garment deformations. This enhancement compels the network to focus on learning spectral information predominantly within the frequency band associated with intricate deformations. Furthermore, building upon standard blend skinning techniques, we introduce target-aware temporal skinning weights. The weights describe how the underlying human skeleton dynamically affects the mesh vertices according to the garment and body shape, as well as the motion state. We validate our method on various garments, bodies, and motions through extensive ablation studies. Finally, we conduct comparisons to confirm its superiority in generalization, deformation quality, and performance over several state-of-the-art methods.

Despite significant advancements in neural simulation techniques for clothing animation, these methods struggle to capture the dynamic details of garments during movement. This limitation restricts their applicability in scenarios where high-quality garment deformation is essential. To address this challenge, we introduce a novel graph learning-based approach to enhance deformation realism through designed mechanisms for mesh information propagation and external optimization strategies during model training. First, we address the issue of over-smoothing common in conventional graph processing techniques by introducing a flexible message-passing method. This approach effectively manages node interactions within the mesh, thereby improving the expressiveness of the model. Furthermore, acknowledging that uniform model supervision typically neglects high-frequency details during optimization, we analyze the spectral properties of clothing meshes. Based on this analysis, we introduce a frequency-division constraint aligned with the characteristics of different frequency bands, which aids in precisely controlling the generation of details. Our model further integrates self-collision and other physics-aware losses, enabling the learning of generalized and fine-grained dynamic deformations. Extensive evaluations and comparisons demonstrate the effectiveness of our approach, showing notable improvements over existing state-of-the-art solutions.

Our work presents a novel spectrum-inspired learning-based approach for generating clothing deformations with dynamic effects and personalized details. Existing methods in the field of clothing animation are limited to either static behavior or specific network models for individual garments, which hinders their applicability in real-world scenarios where diverse animated garments are required. Our proposed method overcomes these limitations by providing a unified framework that predicts dynamic behavior for different garments with arbitrary topology and looseness, resulting in versatile and realistic deformations. First, we observe that the problem of bias towards low frequency always hampers supervised learning and leads to overly smooth deformations. To address this issue, we introduce a frequency-control strategy from a spectral perspective that enhances the generation of high-frequency details of the deformation. In addition, to make the network highly generalizable and able to learn various clothing deformations effectively, we propose a spectral descriptor to achieve a generalized description of the global shape information. Building on the above strategies, we develop a dynamic clothing deformation estimator that integrates graph attention mechanisms with long short-term memory. The estimator takes as input expressive features from garments and human bodies, allowing it to automatically output continuous deformations for diverse clothing types, independent of mesh topology or vertex count. Finally, we present a neural collision handling method to further enhance the realism of garments. Our experimental results demonstrate the effectiveness of our approach on a variety of free-swinging garments and its superiority over state-of-the-art methods.

We introduce a motion retargeting framework capable of animating characters with distinct skeletal structures using video data. While prior studies have successfully performed motion retargeting between skeletons with different structures, retargeting noisy and unnatural motion data extracted from monocular videos has proved challenging. Addressing this issue, our approach proposes a deep learning framework, retargeting motion data procured from easily accessible monocular videos, to animate characters with diverse skeletal structures. Our approach is aimed at providing support for individual creators in character animation.

Abstract This paper presents a novel learning‐based clothing deformation method to generate rich and reasonable detailed deformations for garments worn by bodies of various shapes in various animations. In contrast to existing learning‐based methods, which require numerous trained models for different garment topologies or poses and are unable to easily realize rich details, we use a unified framework to produce high fidelity deformations efficiently and easily. Specifically, we first found that the fit between the garment and the body has an important impact on the degree of folds. We then designed an attribute parser to generate detail‐aware encodings and infused them into the graph neural network, therefore enhancing the discrimination of details under diverse attributes. Furthermore, to achieve better convergence and avoid overly smooth deformations, we proposed to reconstruct output to mitigate the complexity of the learning task. Experimental results show that our proposed deformation method achieves better performance over existing methods in terms of generalization ability and quality of details.

Recent progress in learning-based methods of garment mesh generation is resulting in increased efficiency and maintenance of reality during the generation process. However, none of the previous works so far have focused on variations in material types based on a parameterized material parameter under static poses. In this work, we propose a learning-based method, GarMatNet, for predicting garment deformation based on the functions of human poses and garment materials while maintaining detailed garment wrinkles. GarMatNet consists of two components: a generally-fitting network for predicting smoothed garment mesh and a locally-detailed network for adding detailed wrinkles based on smoothed garment mesh. We hypothesize that material properties play an essential role in the deformation of garments. Since the influences of material type are relatively smaller than pose or body shape, we employ linear interpolation among different factors to control deformation. More specifically, we apply a parameterized material space based on the mass-spring model to express the difference between materials and construct a suitable network structure with weight adjustment between material properties and poses. The experimental results demonstrate that GarMatNet is comparable to the physically-based simulation (PBS) prediction and offers advantages regarding generalization ability, model size, and training time over the baseline model.

Abstract This paper presents a graph‐learning‐based, powerfully generalized method for automatically generating nonlinear deformation for characters with an arbitrary number of vertices. Large‐scale character datasets with a significant number of poses are normally required for training to learn such automatic generalization tasks. There are two key contributions that enable us to address this challenge while making our network generalized to achieve realistic deformation approximation. First, after the automatic linear‐based deformation step, we encode the roughly deformed meshes by constructing graphs where we propose a novel graph feature representation method with three descriptors to represent meshes of arbitrary characters in varying poses. Second, we design a multi‐resolution graph network (MultiResGNet) that takes the constructed graphs as input, and end‐to‐end outputs the offset adjustments of each vertex. By processing multi‐resolution graphs, general features can be better extracted, and the network training no longer heavily relies on large amounts of training data. Experimental results show that the proposed method achieves better performance than prior studies in deformation approximation for unseen characters and poses.

We introduce an unsupervised GAN-based model for shading photorealistic hair animations. Our model is much faster than previous rendering algorithms and produces fewer artifacts than other neural image translation methods. The main idea is to extend the Cycle-GAN structure to avoid semitransparent hair appearance and to exactly reproduce the interaction of the lights with the scene. We use two constraints to ensure temporal coherence and highlight stability. Our approach outperforms and is computationally more efficient than previous methods.

Feathers are sophisticated skin appendages on bird skin, with massive fiber curves (called barbs) branching out from a shaft. Each barb uses its hooklets (called barbules) to further interlock with each other and form two surfaces. We propose a biological modeling scheme that follows the natural feather development to procedurally reproduce common biological characteristics on outputs. Based on our investigations of biology studies, we chooes to generate pathlines of particles in a velocity field to emulate the helical growth of barb curves inside a cylindrical feather follicle, then apply forward kinematics to pathline curves to mimic the unfurling of a feather after its follicle sheath breaks off. We also develop an optional barb snapping algorithm to mimic the geometric restriction from barbules between barbs. Our modeling scheme can achieve feather growth simulation in 3D rather than 2D space, and it is also the first step to prove that it is feasible to alter macroscopic feather geometry via microscopic barbules, both of these topics are less discussed in the field of CG feather modeling. Because of the high compatibility with biology theories, our scheme is expected to be a better basis for discussing other CG feather topics.

In animation production, animators always spend significant time and efforts to develop quality deformation systems for characters with complex appearances and details. In order to decrease the time spent repetitively skinning and fine-tuning work, we propose an end-to-end approach to automatically compute deformations for new characters based on existing graph information of high-quality skinned character meshes. We adopt the idea of regarding mesh deformations as a combination of linear and nonlinear parts and propose a novel architecture for approximating complex nonlinear deformations. Linear deformations on the other hand are simple and therefore can be directly computed, although not precisely. To enable our network handle complicated graph data and inductively predict nonlinear deformations, we design the graph-attention-based (GAT) block to consist of an aggregation stream and a self-reinforced stream in order to aggregate the features of the neighboring nodes and strengthen the features of a single graph node. To reduce the difficulty of learning huge amount of mesh features, we introduce a dense connection pattern between a set of GAT blocks called “dense module” to ensure the propagation of features in our deep frameworks. These strategies allow the sharing of deformation features of existing well-skinned character models with new ones, which we call densely connected graph attention network (DenseGATs). We tested our DenseGATs and compared it with classical deformation methods and other graph-learning-based strategies. Experiments confirm that our network can predict highly plausible deformations for unseen characters.

Abstract In this paper, we propose a novel data‐driven method that uses a machine learning scheme for formulating fracture simulation with the boundary element method (BEM) as a regression problem. With this method, the crack opening displacement (COD) of every correlation node is predicted at the next frame. In our naive prediction, we design a feature vector directly exploiting stress intensities and toughness at the current frame so that our method predicts the COD at the next frame more reliably. Thus, there is no need to solve the original linear BEM system to calculate displacements. This enables us to propagate crack fronts using the estimated stress intensities. There are existing works that use the machine learning approach to accelerate the speed of traditional physics‐based simulations like smoke and fluid, but our work is the first to incorporate the machine learning scheme into BEM‐based fracture simulations. Our implementation accelerates the acquisition of displacements in linear time over the number of crack fronts at each time step compared with the conventional solution whose time complexity grows exponentially based on the BEM linear system. The databases generated by our method are versatile and can be applied to general situations and different models.

Abstract We propose a data‐driven method to realize high‐quality detailed hair animations in interactive applications like games. By devising an error metric method to evaluate hair animation similarities, we take hair features into consideration as much as possible. We also propose a novel database construction algorithm based on Secondary Motion Graph. Our algorithm can improve the efficiency of such graphs to reduce redundant data and also achieves visually smooth connection of two animation clips while taking into consideration their future motions. The costs for the run‐time process using our Secondary Motion Graph are relatively low, allowing real‐time interactive operations. Copyright © 2016 John Wiley & Sons, Ltd.

Abstract This paper proposes an adaptive rendering technique for ray‐bundle tracing. Ray‐bundle tracing can be done by per‐pixel linked‐list construction on a GPU rasterization pipeline. This rasterization based approach offers significant benefits for the efficient generation of light maps (e.g., hardware acceleration, tessellation, and recycling of shaders used in real‐time graphics). However, it is inapplicable to large and complex scenes due to the limited capacity of the GPU memory because it requires a high‐resolution frame buffer and high‐capacity node buffer for the linked‐lists. In addition, memory overflow can potentially occur on the per‐pixel linked‐list since the memory usage of the lists is usually unknown before the rendering process. We introduce an adaptive tiling technique with memory usage prediction. Our method uses an appropriately tiled frame buffer, thus eliminating almost all of the overflow risks thanks to our adaptive tile subdivision scheme. Using this technique, we are able to render high‐quality light maps of large and complex scenes which cannot be computed using previous ray‐bundle based methods.

Surface deformations based on physically-based simulations are used to represent elastic motions such as human skins or clothes in the field of 3DCG applications. LSM (Lattice Shape Matching) [Rivers and James 2007] has particularly attracted attention as a fast and robust method which achieves elastic-like motions. However, the original LSM deformation method generates far from realistic motions especially when stretching an object, because volume is not preserved.

We introduce an algorithm for computing discrete shortest paths on smooth surface representations. For given source and target points on a surface, an initial shortest path is computed by using the Dijkstra’s algorithm. Based on such an initial curve, locally-refined discrete paths are repeatedly calculated so as to get more accurate shortest path. Our algorithm is simple, easy to implement and much fast because it requires only the Dijkstra’s algorithm to compute the path. Our algorithm is a versatile approach for a wide variety of smooth surface representations.

We propose a new approach for surface reconstruction from any unorganized point set, with a fixed amount of parameters. First, we consider some local approximations by compactly supported radial basis functions (CSRBFs), then blend them by a partition of unity method (PU). In our scheme, we start by selecting CSRBFs centers, then construct partitions, finally parameters are calculated in order to minimize, on each sub-domain, the mean square error with a regularization constraint. We demonstrate the effectiveness of our approach on several models with different points distributions.

Sparse low-degree implicit (SLIM) surface is a recently developed non-conforming surface representation. In this paper, a method for free-form deformation of SLIM surfaces is presented. By employing a Laplacian based mesh deformation technique, a global deformation is realized as movements of the centers of the local function supports. A graph connectivity for defining the Laplacians is simply created using inclusion of other centers in the supports. By following the global deformation according to the movements of the centers, we update each local function and its support size via several times of local least square fittings by using the new positions of the neighboring centers. Since the additional computations for updating local functions are very computationally cheap, the proposed SLIM surface deformation achieves an interactive surface deformation similarly to Laplacian based mesh deformation techniques. Further, because a SLIM surface requires a smaller number of elements than a mesh for representing the same geometrical details, the computational effort for a global Laplacian based deformation is dramatically reduced

Nonphoto realistic rendering (NPR) methods have been proposed for more than ten years in the computer graphic field. Separately, within past few years, several algorithms for rendering surfaces from their points, or point sets directly, were proposed. But so far, no NPR like painterly rendering has been proposed for point sets. In this paper, we propose an interactive point-based painterly rendering algorithm. The interactive frame rate is achieved by using both a point-based approach to represent the geometry of the surface and an image-based approach for the rendering. Our algorithm achieves interactive rates and outperforms mesh-based previously reported results. A by-product of our work, we also provide a faster, higher quality IBPR algorithm. We have showed that using most sophisticated statistical approach, we could improve the rendering quality.

We propose a method of subdivision surface quality assessment by reflection lines on programmable graphics hardware (GPU). Using reflection lines is effective for surface quality assessment because the shapes of these lines are changed according to a slight variance of surface shapes. This fact also implies that reflection lines should be calculated precisely. We introduce an intuitive, fast and robust approach for calculating reflection lines by using a plane light source texture based on a fragment program of recently introduced GPU. In addition, we describe a method of calculating position and normal of subdivision surfaces for each fragment on GPU. As our framework for calculating reflection lines does not depend on the level of subdivision, a precise assessment can be established even for low levels of subdivision polygons.

The authors propose a novel multiresolution-based shape representation for 3D mesh morphing. Our approach does not use combination operations that caused some serious problems in the previous approaches for mesh morphing. Therefore, we can calculate a hierarchical interpolation mesh robustly using two types of subdivision fitting schemes. Our new representation has a hierarchical semiregular mesh structure based on subdivision connectivity. This leads to various advantages including efficient data storage, and easy acquisition of an interpolation mesh with arbitrary subdivision level. We also demonstrate several new features for 3D morphing using multiresolution interpolation meshes.

A new algorithm is proposed for calculating the approximate shortest path on a polyhedral surface. The method mainly uses Dijkstra's algorithm and is based on selective refinement of the discrete graph of a polyhedron. Although the algorithm is an approximation, it has the significant advantages of being fast, easy to implement, high approximation accuracy, and numerically robust. The approximation accuracy and computation time are compared between this approximation algorithm and the extended Chen and Han (1990) (ECH) algorithm that can calculate the exact shortest path for non-convex polyhedra. The approximation algorithm can calculate shortest paths within 0.4% accuracy to roughly 100-1000 times faster than the ECH algorithm in our examples. Two applications are discussed of the approximation algorithm to geometric modeling.

This paper proposes a new information filtering method which is used to bridge academic research and business interests. Sometimes terms which are used in business situations have different meanings from those of academic research. Hence, although business persons want to find academic research which matches the needs of their business interests, search results usually contain many unintended material. In order to remove such unintended results, we propose a new filtering method which uses related documents as reference material.

We present a general method for automatically reconstructing a network of B-spline patches based on the Doo-Sabin subdivision surface. This method consists of two parts, surface fitting and surface construction. In surface fitting, mesh simplification based on QEM (Quadric Error Metrics) is used and a control mesh that approximates a Doo-Sabin subdivision surface is constructed. In surface construction, we define a B-spline surface using a surface spline method; and a constructed network of B-spline patches is guaranteed G/sup 1/ continuous. In addition, this method has the advantage of enabling the user to select detail levels of the control mesh by utilizing a mesh simplification process.

The objective of this research is to apply the subdivision surface for surface fitting problems when generating surfaces from data points or polyhedral models. The basic idea is to use the subdivision limit position (SLP) to adapt the control mesh of the subdivision surface to the data points. This method is not a time consuming process involving global optimization. It does, however, fail to capture local characteristics of data points. Consequently, the proposed method is not suitable for generating a surface that precisely interpolates the data points, but it will be useful for quickly generating a surface that captures the overall shape constituted by the data points. A prototype system hers been developed to demonstrate several examples for purposes of evaluating the proposed method.

We propose a 3D mesh dragging method useful for intuitive, efficient geometric modeling of free form polygonal models. With our method, the user can drag a part of a triangular mesh and change its position and orientation. This method is based on an adaptive remeshing procedure which evaluates the deformation of faces by dragging and properly modifies them by deleting or splitting with local topological operations. Therefore the mesh is automatically adjusted for dragging, and irregularity caused by dragging onto the mesh is no longer a concern. In addition, this method is local to the dragged portion, so its computation is efficient. We describe our adaptive remeshing method and demonstrate some dragging examples.

Recently, animations with deforming objects are frequently used in various computer graphics applications. Metamorphosis (or morphing) of three dimensional objects can realize a shape transformation between two or more existing objects. We present a new algorithm for 3D geometric metamorphosis between two objects based on the harmonic map. Our algorithm is applicable for arbitrary polyhedra that are homeomorphic to the three dimensional sphere or the two dimensional disk. In our algorithm, each of the two 3D objects is first embedded to the circular disk on the plane. This embedded model has the same graph structure as its 3D objects. By overlapping those two embedded models, we can establish a correspondence between the two objects. Using this correspondence, intermediate objects between two objects are easily generated. The user only specifies a boundary loop on an object and a vertex on that boundary which control the interpolation.