Gaussian Splatting (GS) has proven to be highly effective in novel view synthesis, achieving high-quality and real-time rendering. However, its potential for reconstructing detailed 3D shapes has not been fully explored. Existing methods often suffer from limited shape accuracy due to the discrete and unstructured nature of Gaussian primitives, which complicates the shape extraction. While recent techniques like 2D GS have attempted to improve shape reconstruction, they often reformulate the Gaussian primitives in ways that reduce both rendering quality and computational efficiency. To address these problems, our work introduces a rasterized approach to render the depth maps and surface normal maps of general 3D Gaussian primitives. Our method not only significantly enhances shape reconstruction accuracy but also maintains the computational efficiency intrinsic to Gaussian Splatting. It achieves a Chamfer distance error comparable to Neuralangelo [33] on the DTU dataset and maintains similar computational efficiency as the original 3D GS methods. Our method is a significant advancement in Gaussian Splatting and can be directly integrated into existing Gaussian Splatting-based methods.
{"title":"RaDe-GS: Rasterizing Depth in Gaussian Splatting","authors":"Baowen Zhang, Chuan Fang, Rakesh Shrestha, Yixun Liang, Xiao-Xiao Long, Ping Tan","doi":"10.1145/3789201","DOIUrl":"https://doi.org/10.1145/3789201","url":null,"abstract":"Gaussian Splatting (GS) has proven to be highly effective in novel view synthesis, achieving high-quality and real-time rendering. However, its potential for reconstructing detailed 3D shapes has not been fully explored. Existing methods often suffer from limited shape accuracy due to the discrete and unstructured nature of Gaussian primitives, which complicates the shape extraction. While recent techniques like 2D GS have attempted to improve shape reconstruction, they often reformulate the Gaussian primitives in ways that reduce both rendering quality and computational efficiency. To address these problems, our work introduces a rasterized approach to render the depth maps and surface normal maps of general 3D Gaussian primitives. Our method not only significantly enhances shape reconstruction accuracy but also maintains the computational efficiency intrinsic to Gaussian Splatting. It achieves a Chamfer distance error comparable to Neuralangelo [33] on the DTU dataset and maintains similar computational efficiency as the original 3D GS methods. Our method is a significant advancement in Gaussian Splatting and can be directly integrated into existing Gaussian Splatting-based methods.","PeriodicalId":50913,"journal":{"name":"ACM Transactions on Graphics","volume":"15 1","pages":""},"PeriodicalIF":6.2,"publicationDate":"2026-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145986205","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Joel Wretborn, Marcus Schoo, Noh-hoon Lee, Christopher Batty, Alexey Stomakhin
This work addresses the challenges of distributing large physics-based simulations often encountered in the visual effects industry. These simulations, based on partial differential equations, model complex phenomena such as free surface liquids, flames, and explosions, and are characterized by domains whose shapes and topologies evolve rapidly. In this context, we propose a novel partitioning algorithm employing optimal transport —which produces a power diagram—and designed to handle a vast variety of simulation domain shapes undergoing rapid changes over time. Our Power partitioner ensures an even distribution of computational tasks, reduces inter-node data exchange, and maintains temporal consistency, all while being intuitive and artist-friendly. To quantify partitioning quality we introduce two metrics, the surface index and the temporal consistency index, which we leverage in a range of comparisons on real-world film production data, showing that our method outperforms the state of the art in a majority of cases.
{"title":"A practical partitioner for distributed simulations on sparse dynamic domains using optimal transport","authors":"Joel Wretborn, Marcus Schoo, Noh-hoon Lee, Christopher Batty, Alexey Stomakhin","doi":"10.1145/3787521","DOIUrl":"https://doi.org/10.1145/3787521","url":null,"abstract":"This work addresses the challenges of distributing large physics-based simulations often encountered in the visual effects industry. These simulations, based on partial differential equations, model complex phenomena such as free surface liquids, flames, and explosions, and are characterized by domains whose shapes and topologies evolve rapidly. In this context, we propose a novel partitioning algorithm employing <jats:italic toggle=\"yes\">optimal transport</jats:italic> —which produces a power diagram—and designed to handle a vast variety of simulation domain shapes undergoing rapid changes over time. Our <jats:italic toggle=\"yes\">Power partitioner</jats:italic> ensures an even distribution of computational tasks, reduces inter-node data exchange, and maintains temporal consistency, all while being intuitive and artist-friendly. To quantify partitioning quality we introduce two metrics, the surface index and the temporal consistency index, which we leverage in a range of comparisons on real-world film production data, showing that our method outperforms the state of the art in a majority of cases.","PeriodicalId":50913,"journal":{"name":"ACM Transactions on Graphics","volume":"3 1","pages":""},"PeriodicalIF":6.2,"publicationDate":"2026-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145968368","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Designing a quad mesh that meets aesthetic, anatomical, and numerical requirements often requires meticulous manual effort in conventional methods, making quadrilateral remeshing an “art of design”. Neural networks hold significant promise for automating this process. However, current approaches that directly predict cross fields cannot properly handle the discontinuous behavior of smooth cross fields: minor shape variations can lead to substantial changes in the cross field, even when singularities remain largely unchanged. Therefore, such methods often result in non-smooth outputs when combining multiple singularity instances. To avoid such discontinuity, we propose to learn the sparse singularities, including their locations and indices, then let the non-neural conventional method to smoothly connect them. The imbalanced ratio of singular and regular vertices poses a significant challenge for learning. We convert them into a geodesic distance field and an over-sampled index field to address it. This carefully designed two-stage strategy satisfies several key requirements, such as coordinate invariance and tessellation insensitivity, while enabling the generation of smooth cross fields with varying topologies. By shifting the focus from directly learning the cross field to learning singularities, we also simplify the dataset preparation process by requiring only sparse annotations.
{"title":"Learning Sparse Singularities for Cross Field Design","authors":"Xiaohu Zhang, Hujun Bao, Jin Huang","doi":"10.1145/3787520","DOIUrl":"https://doi.org/10.1145/3787520","url":null,"abstract":"Designing a quad mesh that meets aesthetic, anatomical, and numerical requirements often requires meticulous manual effort in conventional methods, making quadrilateral remeshing an “art of design”. Neural networks hold significant promise for automating this process. However, current approaches that directly predict cross fields cannot properly handle the discontinuous behavior of smooth cross fields: minor shape variations can lead to substantial changes in the cross field, even when singularities remain largely unchanged. Therefore, such methods often result in non-smooth outputs when combining multiple singularity instances. To avoid such discontinuity, we propose to learn the sparse singularities, including their locations and indices, then let the non-neural conventional method to smoothly connect them. The imbalanced ratio of singular and regular vertices poses a significant challenge for learning. We convert them into a geodesic distance field and an over-sampled index field to address it. This carefully designed two-stage strategy satisfies several key requirements, such as coordinate invariance and tessellation insensitivity, while enabling the generation of smooth cross fields with varying topologies. By shifting the focus from directly learning the cross field to learning singularities, we also simplify the dataset preparation process by requiring only sparse annotations.","PeriodicalId":50913,"journal":{"name":"ACM Transactions on Graphics","volume":"34 1","pages":""},"PeriodicalIF":6.2,"publicationDate":"2026-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145955247","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this work, we propose a system that covers the complete workflow for achieving controlled authoring and editing of textures that present distinctive local characteristics. These include various effects that change the surface appearance of materials, such as stains, tears, holes, abrasions, discoloration, and more. Such alterations are ubiquitous in nature, and including them in the synthesis process is crucial for generating realistic textures. We introduce a novel approach for creating textures with such blemishes, adopting a learning-based approach that leverages unlabeled examples. Our approach does not require manual annotations by the user; instead, it detects the appearance-altering features through unsupervised anomaly detection. The various textural features are then automatically clustered into semantically coherent groups, which are used to guide the conditional generation of images. Our pipeline as a whole goes from a small image collection to a versatile generative model that enables the user to interactively create and paint features on textures of arbitrary size. Notably, the algorithms we introduce for diffusion-based editing and infinite stationary texture generation are generic and should prove useful in other contexts as well. Project page: reality.tf.fau.de/pub/ardelean2025examplebased.html
{"title":"Example-Based Feature Painting on Textures","authors":"Andrei-Timotei Ardelean, Tim Weyrich","doi":"10.1145/3763301","DOIUrl":"https://doi.org/10.1145/3763301","url":null,"abstract":"In this work, we propose a system that covers the complete workflow for achieving controlled authoring and editing of textures that present distinctive local characteristics. These include various effects that change the surface appearance of materials, such as stains, tears, holes, abrasions, discoloration, and more. Such alterations are ubiquitous in nature, and including them in the synthesis process is crucial for generating realistic textures. We introduce a novel approach for creating textures with such blemishes, adopting a learning-based approach that leverages unlabeled examples. Our approach does not require manual annotations by the user; instead, it detects the appearance-altering features through unsupervised anomaly detection. The various textural features are then automatically clustered into semantically coherent groups, which are used to guide the conditional generation of images. Our pipeline as a whole goes from a small image collection to a versatile generative model that enables the user to interactively create and paint features on textures of arbitrary size. Notably, the algorithms we introduce for diffusion-based editing and infinite stationary texture generation are generic and should prove useful in other contexts as well. Project page: reality.tf.fau.de/pub/ardelean2025examplebased.html","PeriodicalId":50913,"journal":{"name":"ACM Transactions on Graphics","volume":"1 1","pages":""},"PeriodicalIF":6.2,"publicationDate":"2025-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145673761","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yiwen Ju, Qingnan Zhou, Xingyi Du, Nathan Carr, Tao Ju
Computing the boundary surface of the 3D volume swept by a rigid or deforming solid remains a challenging problem in geometric modeling. Existing approaches are often limited to sweeping rigid shapes, cannot guarantee a watertight surface, or struggle with modeling the intricate geometric features (e.g., sharp creases and narrow gaps) and topological features (e.g., interior voids). We make the observation that the sweep boundary is a subset of the projection of the intersection of two implicit surfaces in a higher dimension, and we derive a characterization of the subset using winding numbers. These insights lead to a general algorithm for any sweep represented as a smooth time-varying implicit function satisfying a genericity assumption, and it produces a watertight and intersection-free surface that better approximates the geometric and topological features than existing methods.
{"title":"Lifted Surfacing of Generalized Sweep Volumes","authors":"Yiwen Ju, Qingnan Zhou, Xingyi Du, Nathan Carr, Tao Ju","doi":"10.1145/3763360","DOIUrl":"https://doi.org/10.1145/3763360","url":null,"abstract":"Computing the boundary surface of the 3D volume swept by a rigid or deforming solid remains a challenging problem in geometric modeling. Existing approaches are often limited to sweeping rigid shapes, cannot guarantee a watertight surface, or struggle with modeling the intricate geometric features (e.g., sharp creases and narrow gaps) and topological features (e.g., interior voids). We make the observation that the sweep boundary is a subset of the projection of the intersection of two implicit surfaces in a higher dimension, and we derive a characterization of the subset using winding numbers. These insights lead to a general algorithm for any sweep represented as a smooth time-varying implicit function satisfying a genericity assumption, and it produces a watertight and intersection-free surface that better approximates the geometric and topological features than existing methods.","PeriodicalId":50913,"journal":{"name":"ACM Transactions on Graphics","volume":"33 1","pages":""},"PeriodicalIF":6.2,"publicationDate":"2025-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145673762","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Recent advances in image acquisition and scene reconstruction have enabled the generation of high-quality structural urban scene geometry, given sufficient site information. However, current capture techniques often overlook the crucial importance of texture quality, resulting in noticeable visual artifacts in the textured models. In this work, we introduce the urban geometry and texture co-capture problem under limited prior knowledge before a site visit. The only inputs are a 2D building contour map of the target area and a safe flying altitude above the buildings. We propose an innovative aerial path planning framework designed to co-capture images for reconstructing both structured geometry and high-fidelity textures. To evaluate and guide view planning, we introduce a comprehensive texture quality assessment system, including two novel metrics tailored for building facades. Firstly, our method generates high-quality vertical dipping views and horizontal planar views to effectively capture both geometric and textural details. A multi-objective optimization strategy is then proposed to jointly maximize texture fidelity, improve geometric accuracy, and minimize the cost associated with aerial views. Furthermore, we present a sequential path planning algorithm that accounts for texture consistency during image capture. Extensive experiments on large-scale synthetic and real-world urban datasets demonstrate that our approach effectively produces image sets suitable for concurrent geometric and texture reconstruction, enabling the creation of realistic, textured scene proxies at low operational cost.
{"title":"Aerial Path Planning for Urban Geometry and Texture Co-Capture","authors":"Weidan Xiong, Bochuan Zeng, Ziyu Hu, Jianwei Guo, Ke Xie, Hui Huang","doi":"10.1145/3763292","DOIUrl":"https://doi.org/10.1145/3763292","url":null,"abstract":"Recent advances in image acquisition and scene reconstruction have enabled the generation of high-quality structural urban scene geometry, given sufficient site information. However, current capture techniques often overlook the crucial importance of texture quality, resulting in noticeable visual artifacts in the textured models. In this work, we introduce the urban <jats:italic toggle=\"yes\">geometry and texture co-capture</jats:italic> problem under limited prior knowledge before a site visit. The only inputs are a 2D building contour map of the target area and a safe flying altitude above the buildings. We propose an innovative aerial path planning framework designed to co-capture images for reconstructing both structured geometry and high-fidelity textures. To evaluate and guide view planning, we introduce a comprehensive texture quality assessment system, including two novel metrics tailored for building facades. Firstly, our method generates high-quality vertical dipping views and horizontal planar views to effectively capture both geometric and textural details. A multi-objective optimization strategy is then proposed to jointly maximize texture fidelity, improve geometric accuracy, and minimize the cost associated with aerial views. Furthermore, we present a sequential path planning algorithm that accounts for texture consistency during image capture. Extensive experiments on large-scale synthetic and real-world urban datasets demonstrate that our approach effectively produces image sets suitable for concurrent geometric and texture reconstruction, enabling the creation of realistic, textured scene proxies at low operational cost.","PeriodicalId":50913,"journal":{"name":"ACM Transactions on Graphics","volume":"1 1","pages":""},"PeriodicalIF":6.2,"publicationDate":"2025-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145673765","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Julien Philip, Li Ma, Pascal Clausen, Wenqi Xian, Ahmet Levent Taşel, Mingming He, Xueming Yu, David M. George, Ning Yu, Oliver Pilarski, Paul Debevec
We present a unique system for large-scale, multi-performer, high resolution 4D volumetric capture providing realistic free-viewpoint video up to and including 4K resolution facial closeups. To achieve this, we employ a novel volumetric capture, reconstruction and rendering pipeline based on Dynamic Gaussian Splatting and Diffusion-based Detail Enhancement. We design our pipeline specifically to meet the demands of high-end media production. We employ two capture rigs: the Scene Rig , which captures multi-actor performances at a resolution which falls short of 4K production quality, and the Face Rig , which records high-fidelity single-actor facial detail to serve as a reference for detail enhancement. We first reconstruct dynamic performances from the Scene Rig using 4D Gaussian Splatting, incorporating new model designs and training strategies to improve reconstruction, dynamic range, and rendering quality. Then to render high-quality images for facial closeups, we introduce a diffusion-based detail enhancement model. This model is fine-tuned with high-fidelity data from the same actors recorded in the Face Rig. We train on paired data generated from low- and high-quality Gaussian Splatting (GS) models, using the low-quality input to match the quality of the Scene Rig , with the high-quality GS as ground truth. Our results demonstrate the effectiveness of this pipeline in bridging the gap between the scalable performance capture of a large-scale rig and the high-resolution standards required for film and media production.
{"title":"Detail Enhanced Gaussian Splatting for Large-Scale Volumetric Capture","authors":"Julien Philip, Li Ma, Pascal Clausen, Wenqi Xian, Ahmet Levent Taşel, Mingming He, Xueming Yu, David M. George, Ning Yu, Oliver Pilarski, Paul Debevec","doi":"10.1145/3763336","DOIUrl":"https://doi.org/10.1145/3763336","url":null,"abstract":"We present a unique system for large-scale, multi-performer, high resolution 4D volumetric capture providing realistic free-viewpoint video up to and including 4K resolution facial closeups. To achieve this, we employ a novel volumetric capture, reconstruction and rendering pipeline based on Dynamic Gaussian Splatting and Diffusion-based Detail Enhancement. We design our pipeline specifically to meet the demands of high-end media production. We employ two capture rigs: the <jats:italic toggle=\"yes\">Scene Rig</jats:italic> , which captures multi-actor performances at a resolution which falls short of 4K production quality, and the <jats:italic toggle=\"yes\">Face Rig</jats:italic> , which records high-fidelity single-actor facial detail to serve as a reference for detail enhancement. We first reconstruct dynamic performances from the <jats:italic toggle=\"yes\">Scene Rig</jats:italic> using 4D Gaussian Splatting, incorporating new model designs and training strategies to improve reconstruction, dynamic range, and rendering quality. Then to render high-quality images for facial closeups, we introduce a diffusion-based detail enhancement model. This model is fine-tuned with high-fidelity data from the same actors recorded in the <jats:italic toggle=\"yes\">Face Rig.</jats:italic> We train on paired data generated from low- and high-quality Gaussian Splatting (GS) models, using the low-quality input to match the quality of the <jats:italic toggle=\"yes\">Scene Rig</jats:italic> , with the high-quality GS as ground truth. Our results demonstrate the effectiveness of this pipeline in bridging the gap between the scalable performance capture of a large-scale rig and the high-resolution standards required for film and media production.","PeriodicalId":50913,"journal":{"name":"ACM Transactions on Graphics","volume":"168 1","pages":""},"PeriodicalIF":6.2,"publicationDate":"2025-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145673767","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Baptiste Genest, Nicolas Bonneel, Vincent Nivoliers, David Coeurjolly
To solve the optimal transport problem between two uniform discrete measures of the same size, one seeks a bijective assignment that minimizes some matching cost. For this task, exact algorithms are intractable for large problems, while approximate ones may lose the bijectivity of the assignment. We address this issue and the more general cases of non-uniform discrete measures with different total masses, where partial transport may be desirable. The core of our algorithm is a variant of the Quicksort algorithm that provides an efficient strategy to randomly explore many relevant and easy-to-compute couplings, by matching BSP trees in loglinear time. The couplings we obtain are as sparse as possible, in the sense that they provide bijections, injective partial matchings or sparse couplings depending on the nature of the matched measures. To improve the transport cost, we propose efficient strategies to merge k sparse couplings into a higher quality one. For k = 64, we obtain transport plans with typically less than 1% of relative error in a matter of seconds between hundreds of thousands of points in 3D on the CPU. We demonstrate how these high-quality approximations can drastically speed-up usual pipelines involving optimal transport, such as shape interpolation, intrinsic manifold sampling, color transfer, topological data analysis, rigid partial registration of point clouds and image stippling.
{"title":"BSP-OT: Sparse transport plans between discrete measures in loglinear time","authors":"Baptiste Genest, Nicolas Bonneel, Vincent Nivoliers, David Coeurjolly","doi":"10.1145/3763281","DOIUrl":"https://doi.org/10.1145/3763281","url":null,"abstract":"To solve the optimal transport problem between two uniform discrete measures of the same size, one seeks a bijective assignment that minimizes some matching cost. For this task, exact algorithms are intractable for large problems, while approximate ones may lose the bijectivity of the assignment. We address this issue and the more general cases of non-uniform discrete measures with different total masses, where partial transport may be desirable. The core of our algorithm is a variant of the Quicksort algorithm that provides an efficient strategy to randomly explore many relevant and easy-to-compute couplings, by matching BSP trees in loglinear time. The couplings we obtain are as sparse as possible, in the sense that they provide bijections, injective partial matchings or sparse couplings depending on the nature of the matched measures. To improve the transport cost, we propose efficient strategies to merge <jats:italic toggle=\"yes\">k</jats:italic> sparse couplings into a higher quality one. For <jats:italic toggle=\"yes\">k =</jats:italic> 64, we obtain transport plans with typically less than 1% of relative error in a matter of seconds between hundreds of thousands of points in 3D on the CPU. We demonstrate how these high-quality approximations can drastically speed-up usual pipelines involving optimal transport, such as shape interpolation, intrinsic manifold sampling, color transfer, topological data analysis, rigid partial registration of point clouds and image stippling.","PeriodicalId":50913,"journal":{"name":"ACM Transactions on Graphics","volume":"30 1","pages":""},"PeriodicalIF":6.2,"publicationDate":"2025-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145673773","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Markov chain Monte Carlo (MCMC) algorithms are indispensable when sampling from a complex, high-dimensional distribution by a conventional method is intractable. Even though MCMC is a powerful tool, it is also hard to control and tune in practice. Simultaneously achieving both rapid local exploration of the state space and efficient global discovery of the target distribution is a challenging task. In this work, we introduce a novel continuous-time MCMC formulation to the computer science community. Generalizing existing work from the statistics community, we propose a novel framework for adjusting an arbitrary family of Markov processes - used for local exploration of the state space only - to an overall process which is invariant with respect to a target distribution. To demonstrate the potential of our framework, we focus on a simple, but yet insightful, application in light transport simulation. As a by-product, we introduce continuous-time MCMC sampling to the computer graphics community. We show how any existing MCMC-based light transport algorithm can be seamlessly integrated into our framework. We prove empirically and theoretically that the integrated version is superior to the ordinary algorithm. In fact, our approach will convert any existing algorithm into a highly parallelizable variant with shorter running time, smaller error and less variance.
{"title":"Jump Restore Light Transport","authors":"Sascha Holl, Gurprit Singh, Hans-Peter Seidel","doi":"10.1145/3763286","DOIUrl":"https://doi.org/10.1145/3763286","url":null,"abstract":"Markov chain Monte Carlo (MCMC) algorithms are indispensable when sampling from a complex, high-dimensional distribution by a conventional method is intractable. Even though MCMC is a powerful tool, it is also hard to control and tune in practice. Simultaneously achieving both rapid <jats:italic toggle=\"yes\">local exploration</jats:italic> of the state space and efficient <jats:italic toggle=\"yes\">global discovery</jats:italic> of the target distribution is a challenging task. In this work, we introduce a novel continuous-time MCMC formulation to the computer science community. Generalizing existing work from the statistics community, we propose a novel framework for <jats:italic toggle=\"yes\">adjusting</jats:italic> an arbitrary family of Markov processes - used for local exploration of the state space only - to an overall process which is invariant with respect to a target distribution. To demonstrate the potential of our framework, we focus on a simple, but yet insightful, application in light transport simulation. As a by-product, we introduce continuous-time MCMC sampling to the computer graphics community. We show how any existing MCMC-based light transport algorithm can be seamlessly integrated into our framework. We prove empirically and theoretically that the integrated version is superior to the ordinary algorithm. In fact, our approach will convert any existing algorithm into a highly <jats:italic toggle=\"yes\">parallelizable</jats:italic> variant with shorter running time, smaller error and less variance.","PeriodicalId":50913,"journal":{"name":"ACM Transactions on Graphics","volume":"12 1","pages":""},"PeriodicalIF":6.2,"publicationDate":"2025-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145673865","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The demand for high-frame-rate rendering keeps increasing in modern displays. Existing frame generation and super-resolution techniques accelerate rendering by reducing rendering samples across space or time. However, they rely on a uniform sampling reduction strategy, which undersamples areas with complex details or dynamic shading. To address this, we propose to sparsely shade critical areas while reusing generated pixels in low-variation areas for neural extrapolation. Specifically, we introduce the Predictive Error-Flow-eXtrapolation Network (EFXNet)-an architecture that predicts extrapolation errors, estimates flows, and extrapolates frames at once. Firstly, EFXNet leverages temporal coherence to predict extrapolation error and guide the sparse shading of dynamic areas. In addition, EFXNet employs a target-grid correlation module to estimate robust optical flows from pixel correlations rather than pixel values. Finally, EFXNet uses dedicated motion representations for the historical geometric and lighting components, respectively, to extrapolate temporally stable frames. Extensive experimental results show that, compared with state-of-the-art methods, our frame extrapolation method exhibits superior visual quality and temporal stability under a low rendering budget.
{"title":"Consecutive Frame Extrapolation with Predictive Sparse Shading","authors":"Zhizhen Wu, Zhe Cao, Yazhen Yuan, Zhilong Yuan, Rui Wang, Yuchi Huo","doi":"10.1145/3763363","DOIUrl":"https://doi.org/10.1145/3763363","url":null,"abstract":"The demand for high-frame-rate rendering keeps increasing in modern displays. Existing frame generation and super-resolution techniques accelerate rendering by reducing rendering samples across space or time. However, they rely on a uniform sampling reduction strategy, which undersamples areas with complex details or dynamic shading. To address this, we propose to sparsely shade critical areas while reusing generated pixels in low-variation areas for neural extrapolation. Specifically, we introduce the Predictive Error-Flow-eXtrapolation Network (EFXNet)-an architecture that predicts extrapolation errors, estimates flows, and extrapolates frames at once. Firstly, EFXNet leverages temporal coherence to predict extrapolation error and guide the sparse shading of dynamic areas. In addition, EFXNet employs a target-grid correlation module to estimate robust optical flows from pixel correlations rather than pixel values. Finally, EFXNet uses dedicated motion representations for the historical geometric and lighting components, respectively, to extrapolate temporally stable frames. Extensive experimental results show that, compared with state-of-the-art methods, our frame extrapolation method exhibits superior visual quality and temporal stability under a low rendering budget.","PeriodicalId":50913,"journal":{"name":"ACM Transactions on Graphics","volume":"21 1","pages":""},"PeriodicalIF":6.2,"publicationDate":"2025-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145673714","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：