Sampling strategies in computer graphics have long been divided between local approaches, which optimize sample distributions independently at each pixel, and global approaches, which use high-dimensional low-discrepancy sequences to ensure uniformity across all dimensions, including adjacent pixels. While global samplers are most-commonly used thanks to generally having better convergence rates, it comes at a cost of very limited user control over the distribution of samples on subspaces, making it difficult to handle common aliasing artifacts. We introduce a novel modular meta-sampler architecture that bridges local and global sampling, allowing integrator designers to employ specialized low-dimensional samplers while still achieving high-dimensional uniformity. Our approach leverages high-dimensional low-discrepancy sequences to orchestrate sample generation by a collection of local samplers operating over power-of-two hierarchical intervals. We demonstrate how existing local sampling techniques, including stratified, blue noise, and dyadic nets sampling, can be reformulated to be used with this framework, enabling hybrid sampling strategies that combine the benefits of both paradigms.
论文检索
输入标题、作者或关键词，从 100,903 篇学术成果中精准定位
A common process in authoring digital scenes for games, films, and virtual environments is for artists to construct 3D geometry that matches a 2D perspective reference image. In the case of the bare-earth terrain, this is typically a manual process since, unlike for trees and buildings, few inverse reconstruction methods currently exist. To address this, we introduce a method for automatically inferring a detailed, consistent, and complete terrain heightmap from a single photographic image. Our initial phase involves extracting camera parameters and a 3D pointmap from the input image, which is then transformed into a heightmap. However, this only recovers the unoccluded portions of the terrain visible from the perspective of the image. The next phase thus entails the generation of plausible occluded regions using a diffusion model trained on terrain elevation data. The entire process is guided by three consistency principles: geomorphological consistency (the features of the occluded terrain resemble the visible portions), hydrological consistency (the river network is uninterrupted and flows reliably), and view consistency (the shape of the rendered terrain accurately matches the input image). We demonstrate that our method obeys these principles, reliably generates terrains across various scales, and integrates with scene authoring workflows.
While yarn-level garments offer rich dynamic details and compelling visual realism compared to triangle-based models, their wide adoption is hindered by the immense computational cost due to the presence of a large number of degrees of freedom (DOFs). This paper proposes a novel simulation framework designed to enhance the performance and stability for numerical simulation of nonlinear, non-convex, and high-resolution knitwear. Our method generalizes the Douglas-Rachford Splitting (DRS) scheme to resolve the non-convex coupling between stretching, shearing, bending, twisting, and contacting at each yarn thread. A key contribution is a nested decomposition strategy that decouples the non-convex variational energy into independent and convex sub-problems. Such convexification improves solver robustness and removes the necessity for frequent line searches. We provide a theoretically grounded strategy for metric selection for each sub-problem, derived from an analysis of the convergence guarantee of DRS. Consequently, our method achieves close-to-optimal convergence along the nonlinear iterations rather than relying on ad-hoc parameter tuning. The paper also clarifies a formal connection between our generalized DRS and the ADMM (Alternating Direction Method of Multipliers) framework, extending the applicability of our analysis to a broader set of constrained dynamics problems. Experimental results demonstrate that our method robustly handles complex knitwear simulation scenes with superior efficiency, stability, and physical fidelity compared to existing methods. With a matrix-free GPU parallelization, our method allows an interactive simulation rate of knitwear of multi-million DOFs.
We present an efficient B-spline finite element method (FEM) for cloth simulation. While higher-order FEM has long promised higher accuracy, its adoption in cloth simulators has been limited by its larger computational costs while generating results with similar visual quality. Our contribution is a full algorithmic pipeline that makes cloth simulation using quadratic B-spline surfaces faster than standard linear FEM in practice while consistently improving accuracy and visual fidelity. Using quadratic B-spline basis functions, we obtain a globally C1-continuous displacement field that supports consistent discretization of both membrane and bending energies, effectively reducing locking artifacts and mesh dependence common to linear elements. To close the performance gap, we introduce a reduced integration scheme that separately optimizes quadrature rules for membrane and bending energies, an accelerated Hessian assembly procedure tailored to the spline structure, and an optimized linear solver based on partial factorization. Together, these optimizations make high-order, smooth cloth simulation competitive at scale, yielding an average 2× speedup over linear FEM in our tests. Extensive experiments demonstrate improved accuracy, wrinkle detail, and robustness, including contact-rich scenarios, relative to linear FEM and recent higher-order approaches. Our method enables realistic wrinkling dynamics across a wide range of material parameters and supports practical garment animation, providing a new promising spatial discretization for high-quality cloth simulation.
While thin shells have ubiquitous applications and have been studied inside and outside computer graphics for decades, there is little consensus on how to best discretize them. We systematically study models for simulating bending of Kirchhoff-Love shells, with the goal of making practical recommendations, backed by careful numerical experiments, of when and how these models should be used. We analyze ten of the most popular discrete bending models in computer graphics for thin-shell simulation, along with new variants that we propose ourselves. We first verify all models on an analytic test benchmark to probe convergence under refinement and mesh-dependence, and then stress-test with a second benchmark that considers behavior at sharp bends. Finally, we test benchmark leaders on a practical suite of challenging large-scale equilibrium and dynamic shell modeling problems, analyzing both full solution behavior and comparative compute costs. We identify leading existing models and their tradeoffs in terms of accuracy and performance. During this analysis we also identify some issues and modeling gaps in the best-performing discrete bending models. We construct new energy model variants to address some of these gaps, as well as formulas and algorithmic tools for their practical simulation, and finally recommend best practices for modeling thin-shell bending.
Simulating heterogeneous multibody systems with both deformable and stiff components remains a challenge for GPU solvers. While Newton-Krylov methods are popular for their matrix-free nature and good GPU compatibility, they often suffer from severe ill-conditioning and slow convergence when handling stiff contacts and material disparities in deformable multibody systems. In this paper, we present Heterogeneous Subspace Corrections (HSC), a novel Newton-CG variant to efficiently simulate such complex systems. HSC decouples the system into two Krylov iterations: a Newton-CG procedure for deformable bodies and a Neumann-based iteration for the affine subsystem. We introduce a GPU-based Adaptive Cross Approximation (ACA) algorithm to exploit the low-rank nature of the coupling matrices, which manages to reduce the overhead of sparse matrix-vector multiplications substantially on the GPU. For the affine subsystem, we propose a dedicated data structure for fast assembling contact Hessians in parallel. A variety of experimental results demonstrate that our framework consistently outperforms existing GPU simulators, achieving convergence rates comparable to direct solvers even in scenes with high-resolution models and extensive stiff contacts.
The Hong Kong University of Science and Technology (Guangzhou), China and The Hong Kong University of Science and Technology, China Establishing one-to-one stroke correspondences is fundamental to vector-based animation inbetweening. Animators may face great challenges when handling occlusion, as occluded strokes must be drawn explicitly in keyframes and manually hidden frame by frame after stroke interpolation. To reduce tedious effort, we present LayerInbetween, an occlusion-aware framework for vector stroke correspondence and automatic inbetweening. It performs automatic layering to guide stroke tracing and correspondence finding for occluded strokes, and to resolve occlusion with layers in the inbetween frames. To predict occluded strokes, we propose a Global-Local Layer Transformation (GLLT) module that progressively improves the spatial alignment of strokes across keyframes via layer guidance, thereby indicating their potential positions. Our framework is trained on a synthetic dataset comprising 17k+ pairs of keyframes with occlusion and their stroke correspondences. Extensive experiments demonstrate the effectiveness of LayerInbetween compared with existing methods and its generalization capabilities to various types of drawings. In addition to its superior performance, our vector-based inbetweening method enables more flexible editing of 2D animation than raster-based video generation. Code and data for this paper are available at https://github.com/MarkMoHR/LayerInbetween.
Retargeting human kinematic reference motion onto a robot's morphology remains a formidable challenge. Existing methods often produce physical inconsistencies, such as foot sliding, self-collisions, or dynamically infeasible motions, which hinder downstream imitation learning. We propose a bilevel optimization framework that jointly adapts reference motions to a robot's morphology while training a tracking policy using reinforcement learning. To make the optimization tractable, we derive an approximate gradient for the upper-level loss. Our framework requires only a sparse set of semantic rigid-body correspondences and eliminates the need for manual tuning by identifying optimal values for a parameterization expressive enough to preserve characteristic motion across different embodiments. Moreover, by integrating retargeting directly with physics simulation, we produce physically plausible motions that facilitate robust imitation learning. We validate our method in simulation and on hardware, demonstrating challenging motions for morphologies that differ significantly from a human, including retargeting onto a quadruped.
Retargeting motion across characters with varying body shapes while preserving interaction semantics, such as self-contact and near-body proximity, remains a challenging problem. While recent geometry-aware approaches address this by maintaining spatial relationships between predefined corresponding regions, their reliance on static correspondences often struggles when the target character exhibits exaggerated body proportions. In this paper, we present a geometry-aware motion retargeting framework that preserves interaction semantics by performing proximity matching over spatially adaptive anchors. Unlike prior methods with static anchor definitions, the proposed method dynamically repositions anchors to reachable regions on the target character. This is achieved via a Transformer-based anchor refinement strategy that predicts anchor displacements and constrains the translated anchors to remain on the target character geometry through differentiable soft projection. By incorporating pose-dependent spatial structures from the source character, the adapted anchors provide structurally coherent guidance for interaction-aware retargeting. Conditioned on these anchors, a graph-based autoencoder predicts target skeletal motion that preserves the spatial configuration of the source. To encourage task-aligned optimization between anchor adaptation and motion retargeting, we adopt an alternating training scheme in which each module is optimized in turn. Through extensive evaluations, we demonstrate that our method outperforms state-of-the-art approaches in preserving interaction fidelity across diverse character geometries. Code is available at Project Page.
Supporting a wide variety of motion styles is critical for creating diverse virtual characters, but current methods either require large stylized datasets or pre-trained models that cannot generalize beyond their training distribution. We present STyMo, a few-shot approach that learns motion style from only seconds of paired data and trains in one to two minutes. Our key insight is to decompose style into two components: a static component capturing time-invariant posture, and a temporal component capturing frame-wise dynamics. This decomposition yields an interpretable system where posture intensity, temporal exaggeration, and per-body-region style can be adjusted at runtime. Furthermore, the reduction in required training data and computation time structurally permits an iterative authoring workflow. To ensure robustness on arbitrary inputs, we further introduce a stylizability gate that automatically prevents artifacts on out-of-distribution motions. We demonstrate results across diverse motion styles, from subtle emotional variations to exaggerated character archetypes, and release our processed paired dataset to facilitate future research. The source code used in this paper can be found at: https://github.com/facebookresearch/STyMo
In this paper, we study an under-explored but important factor of diffusion generative models, i.e., the combinatorial complexity. Data samples are generally high-dimensional, and for various structured generation tasks, additional attributes are combined to associate with data samples. We show that the space spanned by the combination of dimensions and attributes can be insufficiently covered by existing training schemes of diffusion generative models, potentially limiting test time performance. We present a simple fix to this problem by constructing stochastic processes that fully exploit the combinatorial structures, hence the name ComboStoc. Using this simple strategy, we show that network training is significantly accelerated across diverse data modalities, including images and 3D structured shapes. Moreover, ComboStoc enables a new way of test time generation which uses asynchronous time steps for different dimensions and attributes, thus allowing for varying degrees of control over them. Our code is available at: https://github.com/Xrvitd/ComboStoc.
We propose a complete pipeline for modeling and rendering realistic bird feathers from a single photograph, achieving both high visual fidelity and practical efficiency. Given a single input image of a feather, our approach extracts the feather's shaft curve, outline, and albedo, then reconstructs a compact hierarchical representation in a planar/curve (UV) domain. This representation encodes fine barb and barbule details procedurally, enabling continuous multiscale rendering with correct self-shadowing and masking. We analyze the appearance phenomena of different feathers and propose a new feather scattering model for non-iridescent feathers (e.g., parrot feathers), while introducing an additional sheen lobe to capture the distinctive fluffy rim-lighting effect. Our pipeline produces consistent, realistic results under arbitrary lighting and viewing conditions, and achieves real-time performance with a minimal memory footprint (0.02% of explicit-fiber geometry models), making it a practical solution for digital feather rendering without compromising realism.
In this paper, we propose an efficient method for hair and fur rendering that approximates multiple scattering as volumetric light transport, achieving near path-traced visual fidelity at a drastically reduced cost. Multiple scattering among hair fibers is crucial for a realistic appearance, especially for light-colored hair, but it is extremely expensive to simulate directly. Prior approximations, such as dual scattering and fur BSSRDF, improve performance but often fail for dense hair/fur, yielding an overly dry or blurred appearance. Unlike previous volumetric hair rendering solutions, we treat dense hair/fur as a highly anisotropic participating medium to capture soft volumetric illumination, while still using explicit fiber geometry for direct lighting to preserve fine details. We accumulate hair fiber coverage in screen space and stochastically sample a single-scattering event to approximate higher-order scattering. Our method reproduces the rich appearance of hair and fur resulting from multiple scattering, while running about 7–10× faster than path tracing, making it suitable for use in production. We also demonstrate that our approach is robust under a variety of lighting conditions.
Hair is a rich medium of visual and cultural expression, yet its digital modeling remains challenging due to the duality of fluidity and structure. Many existing generative approaches rely primarily on continuous diffusion fields, which entangle global topology with local texture and obscure the semantic and structural organization of hairstyles. To address this, we propose HairGPT, a strand-centric framework that treats strands as generative primitives and formulates realistic 3D hairstyle synthesis as a dual-decoupled autoregressive sequence modeling problem. Our method applies spatial decoupling across semantic scalp regions and structural decoupling along a hierarchical strand representation, progressing from global layout to fine-grained style. We further introduce a geometric tokenizer and region-aware semantic annotations to guide strand-level generation, enabling compositional editing, synthesis of rare and complex hairstyles, and adaptation to stylized domains. By aligning generative modeling with the workflow of digital grooming, HairGPT turns hair generation from opaque texture synthesis into a structured and semantically controllable authoring process, supporting robust semantic conditioning and high-fidelity results across realistic and stylized domains.
Real-time path tracing increasingly operates under extremely low sampling budgets, often below one sample per pixel, as rendering complexity, resolution, and frame-rate requirements continue to rise. Superresolution is widely used in production because it reduces path-tracing cost by tracing rays on a coarser image grid and reconstructing missing details. This creates a uniform tradeoff between cost and spatial detail: every image region receives the same reduced ray budget, although path-tracing noise, reconstruction difficulty, and perceptual importance vary strongly across the image. Adaptive sampling offers a compelling alternative, but existing end-to-end approaches rely on approximations that break down in sparse regimes. We introduce an end-to-end adaptive sampling and denoising pipeline explicitly designed for the sub-1-spp regime. Our method uses a stochastic formulation of sample placement that enables gradient estimation despite discrete sampling decisions, allowing stable training of a neural sampler at low sampling budgets. To better align optimization with human perception, we propose a tone-mapping-aware training pipeline that integrates differentiable filmic operators and a state-of-the-art perceptual loss, preventing oversampling of regions with low visual impact. In addition, we introduce a gather-based pyramidal denoising filter and a learnable generalization of albedo demodulation tailored to sparse sampling. Our results show consistent improvements over uniform sparse sampling, with notably better reconstruction of perceptually critical details such as specular highlights and shadow boundaries, and demonstrate that adaptive sampling remains effective in the sub-1-spp regime.
We present a novel spatio-temporal reuse framework for time-resolved light transport, enabling efficient Monte Carlo rendering of time-of-flight (ToF) phenomena such as time-gated imaging and transient light capture. Existing ToF rendering methods are computationally expensive, scale poorly to complex dynamic scenes, and are therefore unsuitable for applications with strict latency constraints. To address this limitation, we draw inspiration from ReSTIR, a reuse-based technique for steady-state real-time rendering, and adapt its core principles to interactive-rate ToF simulation. However, naively applying existing ReSTIR methods to ToF rendering leads to severe inefficiency, as reused paths frequently violate optical path-length constraints and thus contribute little or no signal. We overcome this challenge by introducing a path reuse formulation that explicitly enforces physically valid optical path lengths. The key idea is path-length-aware shift mapping, a geometric transformation based on Newton's method that adjusts reused light paths to satisfy temporal gating constraints, inspired by specular manifold exploration in steady-state caustics rendering. The resulting framework substantially improves the efficiency of ToF rendering across a wide range of scenarios, including complex scenes with glossy or specular materials and dynamic motion. Our method supports both time-gated and transient rendering at interactive frame rates, enabling simulation under practical latency constraints. We demonstrate the effectiveness of our approach through two downstream applications, including shape reconstruction and navigation.
We introduce NeuBase, a neural parametric surface representation that both accurately fits target surfaces with fine geometric detail and supports intuitive real time surface deformation. NeuBase consists of a Catmull-Clark subdivision base surface and an offset field defined by a set of neural basis functions encoded via a neural map. By construction, NeuBase surfaces exhibit four fundamental geometric properties, i.e., linearity, locality, smoothness, and affine equivariance, enabling real-time, direct manipulation without retraining the neural network. In addition, we propose a scalable neural map that maintains memory efficiency even for complex shapes with dense control meshes. Experiments on a large-scale dataset demonstrate that our method achieves better fitting accuracy than state-of-the-art neural parametric surface representations.
Freeform surfaces are popular in architectural design for their expressive aesthetics, yet they remain challenging to fabricate. A common strategy is to approximate them with assemblies of planar polygonal panels in a process we term panelization, enabling cost-effective fabrication from sheet materials such as wood, metal, or glass. Our goal is to compute panelizations that are both geometrically faithful and structurally robust. A two-stage pipeline, i.e. geometric panelization followed by structural optimization, can be effective, but joint optimization may yield better outcomes by letting mechanics influence the panelization itself. The challenges are that the panelization task is inherently combinatorial, and structural objectives depend on stresses that have a complex, global dependence on the geometry and are costly to evaluate. Moreover, improving structural performance must be balanced against preserving the input surface's design intent. To address these challenges, we formulate panelization as a continuous surface flow coupled to stress analysis, by modeling candidates as creased elastic shells. We define a global objective by integrating an arbitrary function of stress over the shell and derive efficient gradients with respect to the surface shape to enable optimization. We further introduce flow terms that better preserve the input shape and show that running BFGS in a Sobolev space substantially outperforms standard BFGS and Sobolev gradient descent. We evaluate our method on architectural models, showing that incorporating stress during panelization yields substantial stress reductions and more robust designs compared to optimizing stress after fixing the panelization.
Generating realistic 3D human motions in real-time within interactive applications is key for animation, simulation, and humanoid robotics. While recent offline motion generation approaches offer precise control via text and kinematic constraints, they lack the inference speed required for interactive settings. Conversely, existing online methods enable real-time synthesis but often sacrifice controllability or struggle with complex text semantics and long-horizon goals due to limited context windows. In this work, we introduce ARDY, a streaming generation framework that bridges this gap by enabling high-fidelity motion generation controllable via online text prompts and flexible kinematic constraints. ARDY employs a hybrid representation that combines explicit root features with a latent body embedding, balancing precise trajectory control with efficient generative learning. We propose a two-stage autoregressive transformer denoiser that features variable history context and supports conditioning on flexible, long-horizon kinematic constraints. By training on a large-scale motion capture dataset and being directly conditioned on text labels and kinematic constraints sampled from ground truth poses, ARDY natively learns controllable generation that supports online prompting and flexible long-horizon goals. Extensive evaluations on the HumanML3D benchmark and the large-scale, high-fidelity Bones Rigplay dataset demonstrate ARDY's high motion quality and constraint adherence, validating the efficacy of our key architectural decisions. Finally, we demonstrate the method's practical versatility through an interactive demo featuring dynamic text control, diverse keyframe pose constraints, path following, and interactive locomotion control via mouse and keyboard. Supplementary video results, code, and model releases can be found at https://research.nvidia.com/labs/sil/projects/ardy/.