Abstract: Systems and methods are provided that learn garment deformations such that they are essentially collision free. A diffused, volumetric body model representation of the underlying body together with the construction of a subspace for the garment model that yields a differentiable, canonical space configuration. This subspace is used for the regression of the garment model deformation and its dynamics. In this way, garment model deformations are predicted avoiding collisions, and the complexity for inference is reduced, such that a learned representation yields higher quality than previously achievable. The generated garments exhibit a large amount of spatial and temporal detail, and can be produced extremely quickly via the pre-trained networks.

Abstract: According to various embodiments of the present invention, an optical capture system is provided. In one embodiment, a micro-scale optical capturing system is provided with low divergence (approximately 1°) of the incident light and low acceptance angle (<8°) of the captured light. According to embodiments, a micro-scale optical capturing system is provided with a large number of collimated high-power white LEDs as light sources, between 60 and 100 units, for example, and may be positioned at distances of about 650 mm from the sample. In one embodiment, a digital camera using 50 mm focal objective with a 25 mm length extension tube captures images of the sample. This provides a working distance of approximately 100 mm and at the same time maintains ×0.5 magnification for microscale captures, with an image size of 4×4 microns per pixel.

Abstract: According to various embodiments, an artificial intelligence framework capable of neural synthesis of tileable textures is provided. Using a non-tileable texture input, such as single or multi-layer texture stacks, high-quality tileable textures can be generated that match the appearance of the input samples, while increasing their spatial resolution. Embodiments are provided that leverage a latent space approach in a generative network for synthesizing seamlessly tileable textures, which can maintain semantic consistency in boundaries between tiles. A tileabilty metric is provided as feedback to improve and optimize the tileability of the outout texture, for example using a sampling algorithm that can generate high-resolution, artifact-free tileable textures. In embodiments, a convolutional discriminator is provided for detecting artifacts in the synthesized textures by locally estimating the quality of the synthesized maps.

Abstract: Modeling cross-sections of yarn may include receiving yarn simulation input comprising a descriptive model of a general curvature followed by the yarn, providing a plurality of fibers distributed radially from the center of a ply, setting a base position based on parameters, applying a strain model to simulate the effect of stretch forces applied to the yarn, and outputting a yarn model indicating position and directionality of fibers in the yarn. The technology also relates to real-time modeling of a garment comprising a fabric. For instance, real-time modeling of a garment may include providing an input associated with one or more parameters of the fabric, receiving frames of a computer simulated garment, the computer simulated garment including a simulation of the fabric, the fabric simulation including yarns simulated based on a yarn model.

Abstract: According to various embodiments, a method and system is provided to model geometric properties of garments taking into account the effects of the seams that are created to make the garment for use in computer modeling applications. All the seams in a garment, including overlapping seams and asymetric seamlines, are processed based on geometric operations on 2D models of pieces of fabric from patterns. Working in 2D simplifies folding and stitching. A set of parametric spaces are used to enforce continuity along and across seams. A set of constrained optimizations in 2D and 3D are used to guarantee spacial continuity. Geometries produced can be simulated, providing clothing representations where the mechanical and appearance properties of multi-layer seams are derived automatically by applying geometrical operations and physically-based simulation to fabric pieces.

Abstract: A learning-based clothing animation method and system for highly efficient virtual try-on simulations is provided. Given a garment, the system preprocess a rich database of physically-based dressed character simulations, for multiple body shapes and animations. Then, using a database, the system trains a learning-based model of cloth drape and wrinkles, as a function of body shape and dynamics. A model according to embodiments separates global garment fit, due to body shape, from local garment wrinkles, due to both pose dynamics and body shape. A recurrent neural network is provided to regress garment wrinkles, and the system achieves highly plausible nonlinear effects, in contrast to the blending artifacts suffered by previous methods.

Abstract: According to various embodiments, a framework for transferring micro-scale properties of any kind (geometry, optical properties, material parameters resolution, etc.) of a material, to larger images of the same material is provided. The framework includes a machine learning model capable of learning to perform this transfer from a set of micro-scale images of one or more patches of the material and micro-scale properties of the material. The machine learning framework transfers the micro-scale properties to images of the same material, regardless of its size, resolution, illumination or geometry.

Abstract: Computer implemented method, system and computer program product for simulating the behavior of a knitted fabric at yarn level. The method comprises: retrieving structural information of a knitted fabric; representing each stitch with four contact nodes (4) at the end of the two stitch contacts (5) between pair of loops (2), each contact node (4) being described by a 3D position coordinate (x) and two sliding coordinates (u, v) representing the arc lengths of the two yarns in contact; measuring forces on each contact node (4) based on a force model including wrapping forces to capture the interaction of yarns at stitches; calculating the movement of each contact node (4) at a plurality of time steps using equations of motion derived using the Lagrange-Euler equations, and numerically integrated over time, wherein the equations of motion account for the mass density distributed uniformly along yarns, as well as the measured forces and boundary conditions.

Abstract: A learning-based clothing animation method and system for highly efficient virtual try-on simulations is provided. Given a garment, the system preprocess a rich database of physically-based dressed character simulations, for multiple body shapes and animations. Then, using a database, the system trains a learning-based model of cloth drape and wrinkles, as a function of body shape and dynamics. A model according to embodiments separates global garment fit, due to body shape, from local garment wrinkles, due to both pose dynamics and body shape. A recurrent neural network is provided to regress garment wrinkles, and the system achieves highly plausible nonlinear effects, in contrast to the blending artifacts suffered by previous methods.

Abstract: A learning-based clothing animation method and system for highly efficient virtual try-on simulations is provided. Given a garment, the system preprocess a rich database of physically-based dressed character simulations, for multiple body shapes and animations. Then, using a database, the system trains a learning-based model of cloth drape and wrinkles, as a function of body shape and dynamics. A model according to embodiments separates global garment fit, due to body shape, from local garment wrinkles, due to both pose dynamics and body shape. A recurrent neural network is provided to regress garment wrinkles, and the system achieves highly plausible nonlinear effects, in contrast to the blending artifacts suffered by previous methods.

Abstract: Computer implemented method, system and computer program product for simulating the behavior of a woven fabric at yarn level. The method comprises. retrieving the layout of warp yarns (1), weft yarns (2) and yarn crossing nodes (3): describing each yarn crossing node (3) by a 3D position coordinate (x) and two sliding coordinates, warp sliding coordinate (u) and weft sliding coordinate (v) representing the sliding of warp (1) and weft (2) yarns; measuring forces on each yarn crossing node (3) based on a force model, the forces being measured on both the 3D position coordinate (x) and the sliding coordinates (u, v); calculating the movement of each yarn crossing node (3) using equations of motion derived using the Lagrange-Euler equations, and numerically integrated over time, wherein the equations of motion account for the mass density distributed uniformly along yarns, as well as the measured forces and boundary conditions.

Abstract: According to various embodiments of the present invention, an optical capture system is provided. In one embodiment, a micro-scale optical capturing system is provided with low divergence (approximately 1°) of the incident light and low acceptance angle (<8°) of the captured light. According to embodiments, a micro-scale optical capturing system is provided with a large number of collimated high-power white LEDs as light sources, between 60 and 100 units, for example, and may be positioned at distances of about 650 mm from the sample. In one embodiment, a digital camera using 50 mm focal objective with a 25 mm length extension tube captures images of the sample. This provides a working distance of approximately 100 mm and at the same time maintains ×0.5 magnification for microscale captures, with an image size of 4×4 microns per pixel.

Abstract: The technology relates to modeling cross-sections of yarn. For instance, modeling cross-sections of yarn may include receiving yarn simulation input comprising a descriptive model of a general curvature followed by the yarn, providing a plurality of fibers distributed raidally from the center of a ply, setting a base position based on parameters, applying a strain model to simulate the effect of stretch forces applied to the yarn, and outputting a yarn model indicating position and directionality of fibers in the yarn. The technology also relates to real-time modeling of a garment comprising a fabric. For instance, real-time modeling of a garment may include providing an input associated with one or more parameters of the fabric, receiving frames of a computer simulated garment, the computer simulated garment including a simulation of the fabric, the fabric simulation including yarns simulated based on a yarn model.

Abstract: A learning-based clothing animation method and system for highly efficient virtual try-on simulations is provided. Given a garment, the system preprocess a rich database of physically-based dressed character simulations, for multiple body shapes and animations. Then, using a database, the system trains a learning-based model of cloth drape and wrinkles, as a function of body shape and dynamics. A model according to embodiments separates global garment fit, due to body shape, from local garment wrinkles, due to both pose dynamics and body shape. A recurrent neural network is provided to regress garment wrinkles, and the system achieves highly plausible nonlinear effects, in contrast to the blending artifacts suffered by previous methods.

Abstract: Computer models for bodies based on vertex-based models are enriched by adding nonlinear soft-tissue dynamics to the traditional piece-wise rigid meshes. A neural network is provided for real-time nonlinear soft-tissue regression to enrich skinned 3D animated sequences. The neural network is trained to predict 3D offsets from joint angle velocities and accelerations, as well as earlier dynamic components. The per-vertex rigidity is computed and leveraged to obtain a better-behaved minimization problem. A novel autoencoder is also provided for dimensionality reduction of the 3D vertex displacements that represent nonlinear soft-tissue dynamics in 3D mesh sequences.

Abstract: Computer implemented method, system and computer program product for simulating the behavior of a knitted fabric at yarn level. The method comprises: retrieving structural information of a knitted fabric; representing each stitch with four contact nodes (4) at the end of the two stitch contacts (5) between pair of loops (2), each contact node (4) being described by a 3D position coordinate (x) and two sliding coordinates (u, v) representing the arc lengths of the two yarns in contact; measuring forces on each contact node (4) based on a force model including wrapping forces to capture the interaction of yarns at stitches; calculating the movement of each contact node (4) at a plurality of time steps using equations of motion derived using the Lagrange-Euler equations, and numerically integrated over time, wherein the equations of motion account for the mass density distributed uniformly along yarns, as well as the measured forces and boundary conditions.

Abstract: Methods and systems for suggesting merchandise to a shopper are based on a match between a portion of a three-dimensional (3D) avatar representing the shopper that is relevant to the merchandise, and the corresponding portion of a 3D fit avatar used by a designer and/or manufacturer to create the merchandise or associated with the merchandise by the designer and/or manufacturer.

Abstract: Computer implemented method, system and computer program product for simulating the behavior of a knitted fabric at yarn level. The method comprises: retrieving structural information of a knitted fabric; representing each stitch with four contact nodes (4) at the end of the two stitch contacts (5) between pair of loops (2), each contact node (4) being described by a 3D position coordinate (x) and two sliding coordinates (u, v) representing the arc lengths of the two yarns in contact; measuring forces on each contact node (4) based on a force model including wrapping forces to capture the interaction of yarns at stitches; calculating the movement of each contact node (4) at a plurality of time steps using equations of motion derived using the Lagrange-Euler equations, and numerically integrated over time, wherein the equations of motion account for the mass density distributed uniformly along yarns, as well as the measured forces and boundary conditions.