Fig 2 - uploaded by Andrew Selle
Content may be subject to copyright.
Two examples of cloth showing the types of intricate folds and wrinkles we wish to simulate; both would require high resolution simulation. A satin bed sheet (left) shows that larger scale cloth typically has more folds and wrinkles. Intricate detail is also visible on a Greek sculpture (right) sculpted to resemble a light fabric.
Source publication
In this paper we simulate high resolution cloth consisting of up to 2 million triangles which allows us to achieve highly detailed folds and wrinkles. Since the level of detail is also influenced by object collision and self collision, we propose a more accurate model for cloth-object friction. We also propose a robust history-based repulsion/colli...
Contexts in source publication
Context 1
... contrasts with previous cloth papers that simulated relatively few triangles: [7] used 10-40 thousand elements, [8] used 5-38 thousand elements, and [9] mostly considered a few thousand elements but their highest resolution simulation was a very thin ribbon with 80 thousand elements that exhibits bending but no folds or wrinkles. These resolutions cannot resolve or simulate folds and wrinkles at the granularity of Figure 2. Most simulation techniques would fail if resolution were increased because of two problems: robustness and tractability. ...
Context 2
... in the second phase, modified particle velocities are sent to the processors that own the respective particles, and the remaining internal pairs are processed (again in Gauss-Seidel order) independently by the processor that contains them. See Figure 12 for an illustration of this ap- plication strategy. Note that this strategy always ensures the effective parallel ordering is equivalent to some serial Gauss-Seidel ordering. ...
Context 3
... an initial settling period, we change the coefficient of friction on the wardrobe allowing the cloth to fall and form many folds and wrinkles under the effect of wind drag. The cloth in this simulation resembles the photograph of draped cloth in Figure 15 and the satin depicted in Figure 2. This simulation averaged just over 6 minutes per frame. ...
Similar publications
Citations
... Modeling the mechanics, and the numerical computation, of collisions, multi-body contact, and self-contact has a long history. In computer graphics [13,16,22,42,54], collisions and contact must be accounted for to yield realistic animations. Many practical problems exhibit contact effects, such as modeling aortic valves [8] and particulate suspensions in Stokes flow [44]. ...
We present a numerical method to simulate thick elastic curves that accounts for self-contact and container (obstacle) constraints under large deformations. The base model includes bending and torsion effects, as well as inextensibility. A minimizing movements, descent scheme is proposed for computing energy minimizers, under the non-convex inextensibility, self-contact, and container constraints (if the container is non-convex). At each pseudo time-step of the scheme, the constraints are linearized, which yields a convex minimization problem (at every time-step) with affine equality and inequality constraints. First order conditions are established for the descent scheme at each time-step, under reasonable assumptions on the admissible set. Furthermore, under a mild time-step restriction, we prove energy decrease for the descent scheme, and show that all constraints are satisfied to second order in the time-step, regardless of the total number of time-steps taken. Moreover, we give a modification of the scheme that regularizes the inequality constraints, and establish convergence of the regularized solution. We then discretize the regularized problem with a finite element method using Hermite and Lagrange elements. Several numerical experiments are shown to illustrate the method, including an example that exhibits massive amounts of self-contact for a tightly packed curve inside a sphere.
... PBS is a well-studied methodology to generate high-quality garment animation by modeling the interactions between realworld forces and garments according to physical laws [BW98,Pro97,SSIF08,TPBF87]. However, PBS-based methods are usually computation-hungry and sensitive to low-quality garment typologies because of complex collisions and interactions among garment, body, and physical forces. ...
We address the 3D animation of loose‐fitting garments from a sequence of body motions. State‐of‐the‐art approaches treat all body joints as a whole to encode motion features, which usually gives rise to learned spurious correlations between garment vertices and irrelevant joints as shown in Fig. 1. To cope with the issue, we encode temporal motion features in a joint‐wise manner and learn an association matrix to map human joints only to most related garment regions by encouraging its sparsity. In this way, spurious correlations are mitigated and better performance is achieved. Furthermore, we devise the joint‐specific pose space deformation (PSD) to decompose the high‐dimensional displacements as the combination of dynamic details caused by individual joint poses. Extensive experiments show that our method outperforms previous works in most indicators. Moreover, garment animations are not interfered with by artifacts caused by spurious correlations, which further validates the effectiveness of our approach. The code is available at https://github.com/qiji77/JointNet.
... The field of cloth simulation has been of significant interest within the computer graphics community for many years [Baraff and Witkin 1998;Choi and Ko 2005;Nealen et al. 2006]. The use of advanced techniques such as integration methods [Fierz et al. 2011;Harmon et al. 2009;Hauth et al. 2003;Jiang et al. 2017;Selle et al. 2008;Terzopoulos et al. 1987], strain limiting [Ma et al. 2016;Narain et al. 2012; Thomaszewski et al. 2009;Wang et al. 2010b], and yarnlevel simulation [Cirio et al. 2014;Kaldor et al. 2008;Sperl et al. 2022] have resulted in highly realistic simulations. However, these methods require high computational time, which can restrict their use in real-time applications, particularly for high-resolution cloth geometry. ...
While recent AI-based draping networks have significantly advanced the ability to simulate the appearance of clothes worn by 3D human models, the handling of multi-layered garments remains a challenging task. This paper presents a model for draping multi-layered garments that are unseen during the training process. Our proposed framework consists of three stages: garment embedding, single-layered garment draping, and untangling. The model represents a garment independent to its topological structure by mapping it onto the $UV$ map of a human body model, allowing for the ability to handle previously unseen garments. In the single-layered garment draping phase, the model sequentially drapes all garments in each layer on the body without considering interactions between them. The untangling phase utilizes a GNN-based network to model the interaction between the garments of different layers, enabling the simulation of complex multi-layered clothing. The proposed model demonstrates strong performance on both unseen synthetic and real garment reconstruction data on a diverse range of human body shapes and poses.
... Dynamic 3D human modeling has been a long-standing challenge to 3D vision and graphics communities, as it is critical to various applications, such as VR/AR, animation and robot simulation. Traditional methods leverage welldesigned parametric model [2] and physics-based simulation [61,67,22,20] to model the inner human body and deformable outer cloth separately, but they typically demand huge engineering efforts and expensive computational cost. Recently, many learning based methods have been proposed [36,25,44,32,35,11,4,60]; unfortunately, some of these methods can not model fine-grained geometry details beyond inner body, while the others only support frame-wise reconstruction to produce dynamic sequence. ...
Recent progress in 4D implicit representation focuses on globally controlling the shape and motion with low dimensional latent vectors, which is prone to missing surface details and accumulating tracking error. While many deep local representations have shown promising results for 3D shape modeling, their 4D counterpart does not exist yet. In this paper, we fill this blank by proposing a novel Local 4D implicit Representation for Dynamic clothed human, named LoRD, which has the merits of both 4D human modeling and local representation, and enables high-fidelity reconstruction with detailed surface deformations, such as clothing wrinkles. Particularly, our key insight is to encourage the network to learn the latent codes of local part-level representation, capable of explaining the local geometry and temporal deformations. To make the inference at test-time, we first estimate the inner body skeleton motion to track local parts at each time step, and then optimize the latent codes for each part via auto-decoding based on different types of observed data. Extensive experiments demonstrate that the proposed method has strong capability for representing 4D human, and outperforms state-of-the-art methods on practical applications, including 4D reconstruction from sparse points, non-rigid depth fusion, both qualitatively and quantitatively.
... Physics-based simulation. Pioneering studies achieve realistic clothing animations based on geometric constraints [14,29,25], however, they always suffer from instability and high computational cost. In order to make the simulation efficient, research in [20] computes wrinkles by a static solver and adds them on the coarse base mesh. ...
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. To address the challenging issue of predicting deformations influenced by multi-source attributes, we propose three strategies from novel perspectives. 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 output reconstruction to mitigate the complexity of the learning task. Experiment results show that our proposed deformation method achieves better performance over existing methods in terms of generalization ability and quality of details.
... In this model, surfaces in contact interact throughout normal and shear forces, and sliding occurs when the ratio between the shear and normal force reaches a threshold value, called the static friction coefficient, which is independent of the area of contact and depends only on the roughness of the interacting surfaces. While cloth simulators usually rely on an isotropic Coulomb friction model [BFA02;Sel+09], some recent works have explored anisotropic variations of the Coulomb model when simulating interacting rigid bodies and cloth [PTS09; CFW13; Erl+20]. Interestingly, Chen et al. [CFW13] perform extensive cloth-solid experiments and report a few frictional measurements that exhibit either some anisotropic behaviour, or a non-constant friction coefficient (i.e., a nonlinear relationship between the tangential and normal contact forces). ...
Inverse problems arise in various physical domains and solving them from real-world visual observations poses a significant challenge due to the high dimensional nature of the data. Furthermore gathering enough observations that a data driven model can accurately capture the complete distribution of a physical phenomenon is often intractable. In this work we use deep learning to solve inverse problems by applying two basic principles. Deep learning models can be trained using synthetic data generated from physics based simulations. And the employed simulator itself needs to be verified for physical accuracy thus allowing the model to learn the exact physical phenomenon that is desired. To validate the simulator, we introduce rich and compact physical protocols, originally proposed in soft matter physics literature to measure physical parameters. These protocols can be easily replicated in a simulator to test the physical correctness of the model, and the validity of the simulator. We solve the inverse measurement problem of estimating contact friction in soft-bodies which otherwise requires a specialized physics bench and entails tedious acquisition protocols. This makes the prospect of a purely non-invasive, video-based measurement technique particularly attractive. Previous works have shown that such a video-based estimation is feasible for material parameters using deep learning, but this has never been applied to the friction estimation problem which results in even more subtle visual variations. Since acquiring a large dataset for this problem is impractical, we generate it using a frictional contact simulator. As the simulator has been calibrated and verified using controlled experiments, the results are not only visually plausible, but physically-correct enough to match observations made at the macroscopic scale. We propose to our knowledge the first non-invasive measurement network and adjoining synthetic training dataset for estimating cloth friction at contact, for both cloth-hard body and cloth-cloth contacts. We also acquire an extensive dataset of real world experiments for testing. Both the training and test datasets have been made freely available to the community. We also utilize the same protocol for solving the inverse measurement problem of estimating the deformed curvature of a suspended Kirchhoff rod. In order to do such estimation on physical rods, we utilize a deep learning model to visually predict a curvature field from a suspended rod. As creating a dataset from physical rods (even if synthetically constructed), that faithfully covers a representative manifold of deformed curvatures is intractable, we rely on generating such a dataset from a verified simulator. Our work shows a promising way forward for utilizing deep learning models as part of an inversion measurement pipeline.
... This work was pioneered by Baraff and Witkin's [1998] application of implicit time integration to accelerate cloth simulation. Subsequent research proposes many improvements including implicit-explicit methods [Boxerman and Ascher 2004], adaptive remeshing [Grinspun et al. 2002;Li et al. 2018;Narain et al. , 2012, distributed memory parallelism [Selle et al. 2009], positionbased dynamics [Müller et al. 2007], subdivision thin shell element methods [Vetter et al. 2014], multi-grid methods [Tamstorf et al. 2015], and various approaches to incorporating yarn-level dynamics [Kaldor et al. 2008], such as by homogenization [Sperl et al. 2020] or enrichment of a triangle mesh by yarn patches [Casafranca et al. 2020]. ...
We propose a new model and algorithm to capture the high-definition statics of thin shells via coarse meshes. This model predicts global, fine-scale wrinkling at frequencies much higher than the resolution of the coarse mesh; moreover, it is grounded in the geometric analysis of elasticity, and does not require manual guidance, a corpus of training examples, nor tuning of ad hoc parameters. We first approximate the coarse shape of the shell using tension field theory, in which material forces do not resist compression. We then augment this base mesh with wrinkles, parameterized by an amplitude and phase field that we solve for over the base mesh, which together characterize the geometry of the wrinkles. We validate our approach against both physical experiments and numerical simulations, and we show that our algorithm produces wrinkles qualitatively similar to those predicted by traditional shell solvers requiring orders of magnitude more degrees of freedom.
... 3D virtual try-on, the process of fitting a specific clothing item onto a 3D human shape, has attracted increasing attention due to its promising research and commercial value. Recently, researchers' interest has moved from physics-based [2,5,6,42,13,15] or scan-based approaches [37,27,44] to learning-based 3D try-on methods [3,35,31,55,8], dressing a 3D person directly from 2D images and getting rid of costly physics simulation or 3D sensors. However, most of these learning methods [3,35,31] build on the parametric SMPL [29] model and depend on some predefined digital wardrobe [3], limiting their real-world applicability. ...
... 3D virtual try-on, the process of fitting a specific clothing item onto a 3D human shape, has attracted increasing attention due to its promising research and commercial value. Recently, researchers' interest has moved from physics-based [2,5,6,42,13,15] or scan-based approaches [37,27,44] to learning-based 3D try-on methods [3,35,31,55,8], dressing a 3D person directly from 2D images and getting rid of costly physics simulation or 3D sensors. However, most of these learning methods [3,35,31] build on the parametric SMPL [29] model and depend on some predefined digital wardrobe [3], limiting their real-world applicability. ...
Virtual 3D try-on can provide an intuitive and realistic view for online shopping and has a huge potential commercial value. However, existing 3D virtual try-on methods mainly rely on annotated 3D human shapes and garment templates, which hinders their applications in practical scenarios. 2D virtual try-on approaches provide a faster alternative to manipulate clothed humans, but lack the rich and realistic 3D representation. In this paper, we propose a novel Monocular-to-3D Virtual Try-On Network (M3D-VTON) that builds on the merits of both 2D and 3D approaches. By integrating 2D information efficiently and learning a mapping that lifts the 2D representation to 3D, we make the first attempt to reconstruct a 3D try-on mesh only taking the target clothing and a person image as inputs. The proposed M3D-VTON includes three modules: 1) The Monocular Prediction Module (MPM) that estimates an initial full-body depth map and accomplishes 2D clothes-person alignment through a novel two-stage warping procedure; 2) The Depth Refinement Module (DRM) that refines the initial body depth to produce more detailed pleat and face characteristics; 3) The Texture Fusion Module (TFM) that fuses the warped clothing with the non-target body part to refine the results. We also construct a high-quality synthesized Monocular-to-3D virtual try-on dataset, in which each person image is associated with a front and a back depth map. Extensive experiments demonstrate that the proposed M3D-VTON can manipulate and reconstruct the 3D human body wearing the given clothing with compelling details and is more efficient than other 3D approaches.
... Virtual try-on has seen a surge of interest in image generation and computer graphics research fields. 3D graphicsbased try-on methods [2,4,5,16,18,27,36,39,40] usually rely on intensive computational, manual parameter tuning, or extra facilities, making them unaffordable in practical applications. To reduce the expensive deployment cost for virtual try-on system, recent efforts have been focusing on image-based methods [7,9,19,20,23,32,34,44,50,51], which directly transfer a clothing image onto a reference person image. ...
Despite recent progress on image-based virtual try-on, current methods are constraint by shared warping networks and thus fail to synthesize natural try-on results when faced with clothing categories that require different warping operations. In this paper, we address this problem by finding clothing category-specific warping networks for the virtual try-on task via Neural Architecture Search (NAS). We introduce a NAS-Warping Module and elaborately design a bilevel hierarchical search space to identify the optimal network-level and operation-level flow estimation architecture. Given the network-level search space, containing different numbers of warping blocks, and the operation-level search space with different convolution operations, we jointly learn a combination of repeatable warping cells and convolution operations specifically for the clothing-person alignment. Moreover, a NAS-Fusion Module is proposed to synthesize more natural final try-on results, which is realized by leveraging particular skip connections to produce better-fused features that are required for seamlessly fusing the warped clothing and the unchanged person part. We adopt an efficient and stable one-shot searching strategy to search the above two modules. Extensive experiments demonstrate that our WAS-VTON significantly outperforms the previous fixed-architecture try-on methods with more natural warping results and virtual try-on results.
