论文检索

Zhen Han, Mattias Teye, Derek Yadgaroff, Judith Bütepage

The training of high-quality, robust machine learning models for speech-driven 3D facial animation requires a large, diverse dataset of high-quality audio-animation pairs. To overcome the lack of such a dataset, recent work has introduced large pre-trained speech encoders that are robust to variations in the input audio and, therefore, enable the facial animation model to generalize across speakers, audio quality, and languages. However, the resulting facial animation models are prohibitively large and lend themselves only to offline inference on a dedicated machine. In this work, we explore on-device, real-time facial animation models in the context of game development. We overcome the lack of large datasets by using hybrid knowledge distillation with pseudo-labeling. Given a large audio dataset, we employ a high-performing teacher model to train very small student models. In contrast to the pre-trained speech encoders, our student models only consist of convolutional and fully-connected layers, removing the need for attention context or recurrent updates. In our experiments, we demonstrate that we can reduce the memory footprint to up to 3.4 MB and required future audio context to up to 81 ms while maintaining high-quality animations. This paves the way for on-device inference, an important step towards realistic, model-driven digital characters.

Wenbin Song, Heng Zhang 0027, Yang Wang 0063, Xiaopei Liu

该来源暂未提供摘要。

Hewen Xiao, Xiuping Liu, Hang Zhao 0018, Jian Liu, Kai Xu 0004

We introduce a novel design of parallel-jaw grippers drawing inspiration from pin-pression toys. The proposed pin-pression gripper features a distinctive mechanism in which each finger integrates a 2D array of pins capable of independent extension and retraction. This unique design allows the gripper to instantaneously customize its finger's shape to conform to the object being grasped by dynamically adjusting the extension/retraction of the pins. In addition, the gripper excels in in-hand re-orientation of objects for enhanced grasping stability again via dynamically adjusting the pins. To learn the dynamic grasping skills of pin-pression grippers, we devise a dedicated reinforcement learning algorithm with careful designs of state representation and reward shaping. To achieve a more efficient grasp-while-lift grasping mode, we propose a curriculum learning scheme. Extensive evaluations demonstrate that our design, together with the learned skills, leads to highly flexible and robust grasping with much stronger generality to unseen objects than alternatives. We also highlight encouraging physical results of sim-to-real transfer on a physically manufactured pin-pression gripper, demonstrating the practical significance of our novel gripper design and grasping skill. Demonstration videos for this paper are available at https://github.com/siggraph-pin-pression-gripper/pin-pression-gripper-video.

Ziqi Wang 0006, Wenjun Liu, Jingwen Wang 0006, Gabriel Vallat, Fan Shi 0002, Stefana Parascho, Maryam Kamgarpour

该来源暂未提供摘要。

Haolin Lu 0001, Delio Vicini, Wesley Chang, Tzu-Mao Li

Variance reduction techniques are widely used for reducing the noise of Monte Carlo integration. However, these techniques are typically designed with the assumption that the integrand is scalar-valued. Recognizing that rendering and inverse rendering broadly involve vector-valued integrands, we identify the limitations of classical variance reduction methods in this context. To address this, we introduce ratio control variates, an estimator that leverages a ratio-based approach instead of the conventional difference-based control variates. Our analysis and experiments demonstrate that ratio control variables can significantly reduce the mean squared error of vector-valued integration compared to existing methods and are broadly applicable to various rendering and inverse rendering tasks.

Zhimin Fan 0001, Yiming Wang, Chenxi Zhou, Ling-Qi Yan 0001, Yanwen Guo 0001, Jie Guo 0001

该来源暂未提供摘要。

Qingqin Hua, Pascal Grittmann, Philipp Slusallek

Multiple importance sampling (MIS) is a vital component of most rendering algorithms. MIS computes a weighted sum of samples from many different techniques to achieve generalization, that is, to handle a wide range of scene types and lighting effects. A key factor to the performance of MIS is the choice of weighting function. The go-to default - the balance heuristic - performs well in many cases, but prior work has shown that it can yield unsatisfactory results. A number of challenges cause this suboptimal performance, including low-variance techniques, sample correlation, and unknown sampling densities. Prior work has suggested improvements for some of these problems, but a general optimal solution has yet to be found. We propose a general and practical weight correction scheme: We optimize, on-the-fly, a set of correction factors that are multiplied into any baseline MIS heuristic (e.g., balance or power). We demonstrate that this approach yields consistently better equal-time performance on two rendering applications: bidirectional algorithms and resampled importance sampling for direct illumination.

Nicolas Bonneel, David Coeurjolly, Jean-Claude Iehl, Victor Ostromoukhov

Low-discrepancy sequences, and more particularly Sobol' sequences are gold standard for drawing highly uniform samples for quasi-Monte Carlo applications. They produce so-called (t,s)-sequences, that is, sequences of s-dimensional samples whose uniformity is controlled by a non-negative integer quality factor t. The Monte Carlo integral estimator has a convergence rate that improves as t decreases. Sobol' construction in base 2 also provides extremely fast sampling point generation using efficient xor-based arithmetic. Computer graphics applications, such as rendering, often require high uniformity in consecutive 2D projections and in higher-dimensional projections at the same time. However, it can be shown that, in the classical Sobol' construction, only a single 2D sequence of points (up to scrambling), constructed using irreducible polynomials x and x + 1, achieves the ideal t = 0 property. Reusing this sequence in projections necessarily loses high dimensional uniformity. We prove the existence and construct many 2D Sobol' sequences having t = 1 using irreducible polynomials p and p2 + p + 1. They can be readily combined to produce higher-dimensional low discrepancy sequences with a high-quality t = 1, guaranteed in consecutive pairs of dimensions. We provide the initialization table that can be directly used with any existing Sobol' implementation, along with the corresponding generator matrices, for an optimized 692-dimensional Sobol' construction. In addition to guaranteeing the (1, 2)-sequence property for all consecutive pairs, we ensure that t ≤ 4 for consecutive 4D projections up to 215 points.

Zhiqi Li 0004, Ruicheng Wang, Junlin Li, Duowen Chen 0003, Sinan Wang, Bo Zhu 0002

We propose the Epsilon Difference Gradient Evolution (EDGE) method for accurate flow-map calculation on grids via Hermite interpolation without using velocity buffers. Our key idea is to integrate Gradient Evolution for accurate first-order derivatives and a tetrahedron-based Epsilon Difference scheme to compute higher-order derivatives with reduced memory consumption. EDGE achieves O (1) memory usage, independent of flow map length, while maintaining vorticity preservation comparable to buffer-based methods. We validate our methods across diverse vortical flow scenarios, demonstrating up to 90% backward map memory reduction and significant computational efficiency, broadening the applicability of flow-map methods to large-scale and complex fluid simulations.

Duowen Chen 0003, Zhiqi Li 0004, Taiyuan Zhang, Jinjin He, Junwei Zhou 0001, Bart G. van Bloemen Waanders, Bo Zhu 0002

This paper presents a unified compressible flow map framework designed to accommodate diverse compressible flow systems, including high-Mach-number flows (e.g., shock waves and supersonic aircraft), weakly compressible systems (e.g., smoke plumes and ink diffusion), and incompressible systems evolving through compressible acoustic quantities (e.g., free-surface shallow water). At the core of our approach is a theoretical foundation for compressible flow maps based on Lagrangian path integrals, a novel advection scheme for the conservative transport of density and energy, and a unified numerical framework for solving compressible flows with varying pressure treatments. We validate our method across three representative compressible flow systems, characterized by varying fluid morphologies, governing equations, and compressibility levels, demonstrating its ability to preserve and evolve spatiotemporal features such as vortical structures and wave interactions governed by different flow physics. Our results highlight a wide range of novel phenomena, from ink torus breakup to delta wing tail vortices and vortex shedding on free surfaces, significantly expanding the range of fluid systems that flow-map methods can handle.

Yuchen Sun 0002, Junlin Li, Ruicheng Wang, Sinan Wang, Zhiqi Li 0004, Bart G. van Bloemen Waanders, Bo Zhu 0002

We propose Leapfrog Flow Maps (LFM) to simulate incompressible fluids with rich vortical flows in real time. Our key idea is to use a hybrid velocityimpulse scheme enhanced with leapfrog method to reduce the computational workload of impulse-based flow map methods, while possessing strong ability to preserve vortical structures and fluid details. In order to accelerate the impulse-to-velocity projection, we develop a fast matrix-free Algebraic Multigrid Preconditioned Conjugate Gradient (AMGPCG) solver with customized GPU optimization, which makes projection comparable with impulse evolution in terms of time cost. We demonstrate the performance of our method and its efficacy in a wide range of examples and experiments, such as real-time simulated burning fire ball and delta wingtip vortices.

Mengdi Wang 0003, Fan Feng, Junlin Li, Bo Zhu 0002

We propose the adaptive hybrid particle-grid flow map method, a novel flow-map approach that leverages Lagrangian particles to simultaneously transport impulse and guide grid adaptation, introducing a fully adaptive flow map-based fluid simulation framework. The core idea of our method is to maintain flow-map trajectories separately on grid nodes and particles: the grid-based representation tracks long-range flow maps at a coarse spatial resolution, while the particle-based representation tracks both long and short-range flow maps, enhanced by their gradients, at a fine resolution. This hybrid Eulerian-Lagrangian flow-map representation naturally enables adaptivity for both advection and projection steps. We implement this method in Cirrus, a GPU-based fluid simulation framework designed for octree-like adaptive grids enhanced with particle trackers. The efficacy of our system is demonstrated through numerical tests and various simulation examples, achieving up to 512 × 512 × 2048 effective resolution on an RTX 4090 GPU. We achieve a 1.5 to 2× speedup with our GPU optimization over the Particle Flow Map method on the same hardware, while the adaptive grid implementation offers efficiency gains of one to two orders of magnitude by reducing computational resource requirements. The source code has been made publicly available at: https://wang-mengdi.github.io/proj/25-cirrus/.

Zhiqi Li 0004, Candong Lin, Duowen Chen 0003, Xinyi Zhou, Shiying Xiong, Bo Zhu 0002

We propose a novel gauge fluid solver that evolves Clebsch wave functions on particle flow maps (PFMs). The key insight underlying our work is that particle flow maps exhibit superior performance in transporting point elements—such as Clebsch components—compared to line and surface elements, which were the focus of previous methods relying on impulse and vortex gauge variables for flow maps. Our Clebsch PFM method incorporates three main contributions: a novel gauge transformation enabling accurate transport of wave functions on particle flow maps, an enhanced velocity reconstruction method for coarse grids, and a PFM-based simulation framework designed to better preserve fine-scale flow structures. We validate the Clebsch PFM method through a wide range of benchmark tests and simulation examples, ranging from leapfrogging vortex rings and vortex reconnections to Kelvin-Helmholtz instabilities, demonstrating that our method outperforms its impulse- or vortex-based counterparts on particle flow maps, particularly in preserving and evolving small-scale features.

Sinan Wang, Junwei Zhou 0001, Fan Feng, Zhiqi Li 0004, Yuchen Sun 0002, Duowen Chen 0003, Greg Turk, Bo Zhu 0002

We propose the Vortex Particle Flow Map (VPFM) method to simulate incompressible flow with complex vortical evolution in the presence of dynamic solid boundaries. The core insight of our approach is that vorticity is an ideal quantity for evolution on particle flow maps, enabling significantly longer flow map distances compared to other fluid quantities like velocity or impulse. To achieve this goal, we developed a hybrid Eulerian-Lagrangian representation that evolves vorticity and flow map quantities on vortex particles, while reconstructing velocity on a background grid. The method integrates three key components: (1) a vorticity-based particle flow map framework, (2) an accurate Hessian evolution scheme on particles, and (3) a solid boundary treatment for no-through and no-slip conditions in VPFM. These components collectively allow a substantially longer flow map length (3–12 times longer) than the state-of-the-art, enhancing vorticity preservation over extended spatiotemporal domains. We validated the performance of VPFM through diverse simulations, demonstrating its effectiveness in capturing complex vortex dynamics and turbulence phenomena.

Chris Giles, Elie Diaz, Cem Yuksel

Vertex Block Descent is a fast physics-based simulation method that is unconditionally stable, highly parallelizable, and capable of converging to the implicit Euler solution. We extend it using an augmented Lagrangian formulation to address some of its fundamental limitations. First, we introduce a mechanism to handle hard constraints with infinite stiffness without introducing numerical instabilities. Second, we substantially improve the convergence in the presence of high stiffness ratios. These changes we introduce allow simulating complex contact scenarios involving rigid bodies with stacking and friction, articulated bodies connected with hard constraints, including joints with limited degrees of freedom, and stiff systems interacting with soft bodies. We present evaluations using a parallel GPU implementation that can deliver real-time performance and stable simulations with low iteration counts for millions of objects interacting via collisions, various joint/attachment constraints, and springs of various stiffness. Our results show superior performance, convergence, and stability compared to the state-of-the-art alternatives.

Jiahao Wen, Jernej Barbic, Danny M. Kaufman

该来源暂未提供摘要。

Ziqiu Zeng, Siyuan Luo, Fan Shi 0002, Zhongkai Zhang 0001

We present a GPU-friendly framework for real-time implicit simulation of elastic material in the presence of frictional contacts. The integration of hyperelasticity, non-interpenetration contact, and friction in real-time simulations presents formidable nonlinear and non-smooth problems, which are highly challenging to solve. By incorporating nonlinear complementarity conditions within the local-global framework, we achieve rapid convergence in addressing these challenges. While the structure of local-global methods is not fully GPU-friendly, our proposal of a simple yet efficient solver with sparse presentation of the system inverse enables highly parallel computing while maintaining a fast convergence rate. Moreover, our novel splitting strategy for non-smooth indicators not only amplifies overall performance but also refines the complementarity preconditioner, enhancing the accuracy of frictional behavior modeling. Through extensive experimentation, the robustness of our framework in managing real-time contact scenarios, ranging from large-scale systems and extreme deformations to non-smooth contacts and precise friction interactions, has been validated. Compatible with a wide range of hyperelastic models, our approach maintains efficiency across both low and high stiffness materials. Despite its remarkable efficiency, robustness, and generality, our method is elegantly simple, with its core contributions grounded solely on standard matrix operations.

Huanyu Chen, Jiahao Wen, Jernej Barbic

We give a method to simulate large deformations of 3D elastic rods under arbitrary nonlinear isotropic 3D solid materials. Rod elastic energies in existing graphics literature are derived from volumetric models under the small-strain linearization assumptions. While the resulting equations can and are commonly applied to large deformations, the material modeling has been limited to a single material, namely linear Hooke law. Starting from any 3D solid nonlinear isotropic elastic energy density function ψ, we derive our rod elastic energy by subjecting the 3D solid volumetric material to the limit process whereby rod thickness is decreased to zero. This enables us to explain rod stretching, bending and twisting in a unified model. Care must be taken to adequately model cross-sectional in-plane and out-of-plane deformations. Our key insight is to compute the three cross-sectional deformation modes corresponding to bending (in the two directions) and twisting, using linear theory. Then, given any ψ, we use these modes to derive an analytical formula for a 5D "macroscopic" large-deformation rod elastic energy function of the local longitudinal stretch, radial scaling, the two bending curvatures and torsion. Our model matches linear theory for small deformations, including cross-sectional shrinkage due to Poisson's effect, and produces correct bending and torsional constants. Our experiments demonstrate that our energy closely matches volumetric FEM even under large stretches and curvatures, whereas commonly used methods in graphics deviate from it. We also compare to closely related work from mechanics literature; we give an explicit expansion of all energy terms in terms of the rod cross-section diameter, allowing independent adjustment of stretching, bending and twisting. Finally, we observe an inherent limitation in the ability of rod models to control nonlinear bendability and twistability. We propose to "relax" rod physics to more easily control nonlinear bending and twisting in computer graphics applications.

Kaixin Yao, Longwen Zhang, Xinhao Yan, Yan Zeng, Qixuan Zhang, Lan Xu 0003, Wei Yang 0034, Jiayuan Gu, Jingyi Yu 0001

Recovering high-quality 3D scenes from a single RGB image is a challenging task in computer graphics. Current methods often struggle with domain-specific limitations or low-quality object generation. To address these, we propose CAST (Component-Aligned 3D Scene Reconstruction from a Single RGB Image), a novel method for 3D scene reconstruction. CAST starts by extracting object-level 2D segmentation and relative depth information from the input image, followed by using a GPT-based model to analyze inter-object spatial relations. This enables understanding of how objects relate to each other within the scene, ensuring more coherent reconstruction. CAST then employs an occlusion-aware large-scale 3D generation model to independently generate each object's full geometry, using Masked Auto Encoder (MAE) and point cloud conditioning to mitigate the effects of occlusions and partial object information, ensuring accurate alignment with the source image's geometry and texture. To align each object with the scene, the alignment generation model computes the necessary transformations, allowing the generated meshes to be accurately placed and integrated into the scene's point cloud. Finally, CAST applies a physics-aware correction mechanism, which leverages a fine-grained relation graph to generate a constraint graph. This graph guides the optimization of object poses, ensuring physical consistency and spatial coherence. By utilizing Signed Distance Fields (SDF), the model effectively addresses issues such as occlusions, object penetration, and floating objects, ensuring that the generated scene accurately reflects real-world physical interactions. Experimental results demonstrate that CAST significantly improves the quality of single-image 3D scene reconstruction, offering enhanced realism and accuracy in scene understanding and reconstruction tasks. CAST has practical applications in virtual content creation, such as immersive game environments and film production, where real-world setups can be seamlessly integrated into virtual landscapes. Additionally, CAST can be leveraged in robotics, enabling efficient real-to-simulation workflows and providing realistic, scalable simulation environments for robotic systems.

Jianjun Xia, Tao Ju 0001

Surface reconstruction from points is a fundamental problem in computer graphics. While numerous methods have been proposed, it remains challenging to reconstruct from sparse and non-uniform point distributions, particularly when normals are absent. We present a robust and scalable method for reconstructing an implicit surface from points without normals. By exploring the locality of natural neighborhoods, we propose local reformulations of a previous global method, known for its ability to surface sparse points but high computational cost, thereby significantly improving its scalability while retaining its robustness. Experiments show that our method achieves comparable speed to existing reconstruction methods on large inputs while producing fewer artifacts in under-sampled regions.