7 September 2020 in The Visual Computer
1 January 2007 in Computer-Aided Design and Applications
In this paper, we propose a fragment-based evaluation method for non-uniform B-spline surfaces using recent programmable graphics hardware (GPU). A position on a non-uniform B-spline surface is evaluated by the linear combination of both control points and B-spline basis functions. Hence the computational costs can be reduced by pre-computing a knot interval of a parameter from a knot vector. We show that efficient computation of positions and normal vectors for B-spline surfaces can be done on GPUs by applying these ideas. Our algorithm computes an exact position, a derivative and a curvature per fragment. We demonstrate that this achieves high-quality surface rendering. We also discuss about the extension to NURBS and trimmed surfaces.
1 January 2012 in Computer-Aided Design and Applications
A curve skeleton of a 3D object is one of the most important structures of the object, which is extremely useful for many computer graphics applications involving in shape analysis. Much research has focused on volumetric and polygonal mesh models. However, only a few have been paid attention to point cloud and suffered from several limitations such as connectivity or inside-shape information requirements, difficulty to compute junction nodes, or inappropriate parameter tuning. To resolve these problems, we propose an effective algorithm for extracting a robust and centered curve skeleton from point cloud. The process starts from estimating centers of antipodes of a point cloud, so called skeletal candidates, which are fundamentally inside the shape; but scattered. We thus filter and shrink them to create less-noisy skeletal candidates before applying one-dimensional Moving Least Squares (MLS) to build a thin point cloud. The latter is then downsampled to sparse skeletal nodes. With recent nodes, ...
22 December 2023 in IEEE Transactions on Visualization and Computer Graphics
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.
19 May 2025 in IEEE Transactions on Pattern Analysis and Machine Intelligence
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.
1 May 2021 in Computer Graphics Forum