Milin Kodnongbua, Zachary Englhardt, Ricardo Bianchini, Rodrigo Fonseca, Alvin Lebeck, Daniel S. Berger, Vikram Iyer, Fiodar Kazhamiaka, Adriana Schulz
The growing demands for computational power in cloud computing have led to a significant increase in the deployment of high-performance servers. The growing power consumption of servers and the heat they produce is on track to outpace the capacity of conventional air cooling systems, necessitating more efficient cooling solutions such as liquid immersion cooling. The superior heat exchange capabilities of immersion cooling both eliminates the need for bulky heat sinks, fans, and air flow channels while also unlocking the potential go beyond conventional 2D blade servers to three-dimensional designs. In this work, we present a computational framework to explore designs of servers in three-dimensional space, specifically targeting the maximization of server density within immersion cooling tanks. Our tool is designed to handle a variety of physical and electrical server design constraints. We demonstrate our optimized designs can reduce server volume by 25--52% compared to traditional flat server designs. This increased density reduces land usage as well as the amount of liquid used for immersion, with significant reduction in the carbon emissions embodied in datacenter buildings. We further create physical prototypes to simulate dense server designs and perform real-world experiments in an immersion cooling tank demonstrating they operate at safe temperatures. This approach marks a critical step forward in sustainable and efficient datacenter management.
{"title":"Dense Server Design for Immersion Cooling","authors":"Milin Kodnongbua, Zachary Englhardt, Ricardo Bianchini, Rodrigo Fonseca, Alvin Lebeck, Daniel S. Berger, Vikram Iyer, Fiodar Kazhamiaka, Adriana Schulz","doi":"10.1145/3687965","DOIUrl":"https://doi.org/10.1145/3687965","url":null,"abstract":"The growing demands for computational power in cloud computing have led to a significant increase in the deployment of high-performance servers. The growing power consumption of servers and the heat they produce is on track to outpace the capacity of conventional air cooling systems, necessitating more efficient cooling solutions such as liquid immersion cooling. The superior heat exchange capabilities of immersion cooling both eliminates the need for bulky heat sinks, fans, and air flow channels while also unlocking the potential go beyond conventional 2D blade servers to three-dimensional designs. In this work, we present a computational framework to explore designs of servers in three-dimensional space, specifically targeting the maximization of server density within immersion cooling tanks. Our tool is designed to handle a variety of physical and electrical server design constraints. We demonstrate our optimized designs can reduce server volume by 25--52% compared to traditional flat server designs. This increased density reduces land usage as well as the amount of liquid used for immersion, with significant reduction in the carbon emissions embodied in datacenter buildings. We further create physical prototypes to simulate dense server designs and perform real-world experiments in an immersion cooling tank demonstrating they operate at safe temperatures. This approach marks a critical step forward in sustainable and efficient datacenter management.","PeriodicalId":50913,"journal":{"name":"ACM Transactions on Graphics","volume":"22 1","pages":""},"PeriodicalIF":6.2,"publicationDate":"2024-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142673088","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}
Yuchen Sun, Linglai Chen, Weiyuan Zeng, Tao Du, Shiying Xiong, Bo Zhu
This paper introduces a two-phase interfacial fluid model based on the impulse variable to capture complex vorticity-interface interactions. Our key idea is to leverage bidirectional flow map theory to enhance the transport accuracy of both vorticity and interfaces simultaneously and address their coupling within a unified Eulerian framework. At the heart of our framework is an impulse ghost fluid method to solve the two-phase incompressible fluid characterized by its interfacial dynamics. To deal with the history-dependent jump of gauge variables across a dynamic interface, we develop a novel path integral formula empowered by spatiotemporal buffers to convert the history-dependent jump condition into a geometry-dependent jump condition when projecting impulse to velocity. We demonstrate the efficacy of our approach in simulating and visualizing several interface-vorticity interaction problems with cross-phase vortical evolution, including interfacial whirlpool, vortex ring reflection, and leapfrogging bubble rings.
{"title":"An Impulse Ghost Fluid Method for Simulating Two-Phase Flows","authors":"Yuchen Sun, Linglai Chen, Weiyuan Zeng, Tao Du, Shiying Xiong, Bo Zhu","doi":"10.1145/3687963","DOIUrl":"https://doi.org/10.1145/3687963","url":null,"abstract":"This paper introduces a two-phase interfacial fluid model based on the impulse variable to capture complex vorticity-interface interactions. Our key idea is to leverage bidirectional flow map theory to enhance the transport accuracy of both vorticity and interfaces simultaneously and address their coupling within a unified Eulerian framework. At the heart of our framework is an impulse ghost fluid method to solve the two-phase incompressible fluid characterized by its interfacial dynamics. To deal with the history-dependent jump of gauge variables across a dynamic interface, we develop a novel path integral formula empowered by spatiotemporal buffers to convert the history-dependent jump condition into a geometry-dependent jump condition when projecting impulse to velocity. We demonstrate the efficacy of our approach in simulating and visualizing several interface-vorticity interaction problems with cross-phase vortical evolution, including interfacial whirlpool, vortex ring reflection, and leapfrogging bubble rings.","PeriodicalId":50913,"journal":{"name":"ACM Transactions on Graphics","volume":"66 1","pages":""},"PeriodicalIF":6.2,"publicationDate":"2024-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142673097","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}
We develop a computational pipeline to facilitate the biomimetic design of winged seeds. Our approach leverages 3D scans of natural winged seeds to construct a bio-inspired design space by interpolating them with geodesic coordinates in the 3D diffeomorphism group. We formulate aerodynamic design tasks with probabilistic performance objectives and adapt a gradient-free optimizer to explore the design space and minimize the expectation of performance objectives efficiently and effectively. Our pipeline discovers novel winged seed designs that outperform natural counterparts in aerodynamic tasks, including long-distance dispersal and guided flight. We validate the physical fidelity of our pipeline by showcasing paper models of selected winged seeds in the design space and reporting their similar aerodynamic behaviors in simulation and reality.
{"title":"Computational Biomimetics of Winged Seeds","authors":"Qiqin Le, Jiamu Bu, Yanke Qu, Bo Zhu, Tao Du","doi":"10.1145/3687899","DOIUrl":"https://doi.org/10.1145/3687899","url":null,"abstract":"We develop a computational pipeline to facilitate the biomimetic design of winged seeds. Our approach leverages 3D scans of natural winged seeds to construct a bio-inspired design space by interpolating them with geodesic coordinates in the 3D diffeomorphism group. We formulate aerodynamic design tasks with probabilistic performance objectives and adapt a gradient-free optimizer to explore the design space and minimize the expectation of performance objectives efficiently and effectively. Our pipeline discovers novel winged seed designs that outperform natural counterparts in aerodynamic tasks, including long-distance dispersal and guided flight. We validate the physical fidelity of our pipeline by showcasing paper models of selected winged seeds in the design space and reporting their similar aerodynamic behaviors in simulation and reality.","PeriodicalId":50913,"journal":{"name":"ACM Transactions on Graphics","volume":"14 1","pages":""},"PeriodicalIF":6.2,"publicationDate":"2024-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142673117","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}
Hajin Choi, Seokpyo Hong, Inwoo Ha, Nahyup Kang, Bochang Moon
Generating a rendered image sequence through Monte Carlo ray tracing is an appealing option when one aims to accurately simulate various lighting effects. Unfortunately, interactive rendering scenarios limit the allowable sample size for such sampling-based light transport algorithms, resulting in an unbiased but noisy image sequence. Image denoising has been widely adopted as a post-sampling process to convert such noisy image sequences into biased but temporally stable ones. The state-of-the-art strategy for interactive image denoising involves devising a deep neural network and training this network via supervised learning, i.e., optimizing the network parameters using training datasets that include an extensive set of image pairs (noisy and ground truth images). This paper adopts the prevalent approach for interactive image denoising, which relies on a neural network. However, instead of supervised learning, we propose a different learning strategy that trains our network parameters on the fly, i.e., updating them online using runtime image sequences. To achieve our denoising objective with online learning, we tailor local regression to a cross-regression form that can guide robust training of our denoising neural network. We demonstrate that our denoising framework effectively reduces noise in input image sequences while robustly preserving both geometric and non-geometric edges, without requiring the manual effort involved in preparing an external dataset.
{"title":"Online Neural Denoising with Cross-Regression for Interactive Rendering","authors":"Hajin Choi, Seokpyo Hong, Inwoo Ha, Nahyup Kang, Bochang Moon","doi":"10.1145/3687938","DOIUrl":"https://doi.org/10.1145/3687938","url":null,"abstract":"Generating a rendered image sequence through Monte Carlo ray tracing is an appealing option when one aims to accurately simulate various lighting effects. Unfortunately, interactive rendering scenarios limit the allowable sample size for such sampling-based light transport algorithms, resulting in an unbiased but noisy image sequence. Image denoising has been widely adopted as a post-sampling process to convert such noisy image sequences into biased but temporally stable ones. The state-of-the-art strategy for interactive image denoising involves devising a deep neural network and training this network via supervised learning, i.e., optimizing the network parameters using training datasets that include an extensive set of image pairs (noisy and ground truth images). This paper adopts the prevalent approach for interactive image denoising, which relies on a neural network. However, instead of supervised learning, we propose a different learning strategy that trains our network parameters on the fly, i.e., updating them online using runtime image sequences. To achieve our denoising objective with online learning, we tailor local regression to a cross-regression form that can guide robust training of our denoising neural network. We demonstrate that our denoising framework effectively reduces noise in input image sequences while robustly preserving both geometric and non-geometric edges, without requiring the manual effort involved in preparing an external dataset.","PeriodicalId":50913,"journal":{"name":"ACM Transactions on Graphics","volume":"18 1","pages":""},"PeriodicalIF":6.2,"publicationDate":"2024-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142673093","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 paper, we present an implicit surface reconstruction method with 3D Gaussian Splatting (3DGS), namely 3DGSR, that allows for accurate 3D reconstruction with intricate details while inheriting the high efficiency and rendering quality of 3DGS. The key insight is to incorporate an implicit signed distance field (SDF) within 3D Gaussians for surface modeling, and to enable the alignment and joint optimization of both SDF and 3D Gaussians. To achieve this, we design coupling strategies that align and associate the SDF with 3D Gaussians, allowing for unified optimization and enforcing surface constraints on the 3D Gaussians. With alignment, optimizing the 3D Gaussians provides supervisory signals for SDF learning, enabling the reconstruction of intricate details. However, this only offers sparse supervisory signals to the SDF at locations occupied by Gaussians, which is insufficient for learning a continuous SDF. Then, to address this limitation, we incorporate volumetric rendering and align the rendered geometric attributes (depth, normal) with that derived from 3DGS. In sum, these two designs allow SDF and 3DGS to be aligned, jointly optimized, and mutually boosted. Our extensive experimental results demonstrate that our 3DGSR enables high-quality 3D surface reconstruction while preserving the efficiency and rendering quality of 3DGS. Besides, our method competes favorably with leading surface reconstruction techniques while offering a more efficient learning process and much better rendering qualities.
{"title":"3DGSR: Implicit Surface Reconstruction with 3D Gaussian Splatting","authors":"Xiaoyang Lyu, Yang-Tian Sun, Yi-Hua Huang, Xiuzhe Wu, Ziyi Yang, Yilun Chen, Jiangmiao Pang, Xiaojuan Qi","doi":"10.1145/3687952","DOIUrl":"https://doi.org/10.1145/3687952","url":null,"abstract":"In this paper, we present an implicit surface reconstruction method with 3D Gaussian Splatting (3DGS), namely 3DGSR, that allows for accurate 3D reconstruction with intricate details while inheriting the high efficiency and rendering quality of 3DGS. The key insight is to incorporate an implicit signed distance field (SDF) within 3D Gaussians for surface modeling, and to enable the alignment and joint optimization of both SDF and 3D Gaussians. To achieve this, we design coupling strategies that align and associate the SDF with 3D Gaussians, allowing for unified optimization and enforcing surface constraints on the 3D Gaussians. With alignment, optimizing the 3D Gaussians provides supervisory signals for SDF learning, enabling the reconstruction of intricate details. However, this only offers sparse supervisory signals to the SDF at locations occupied by Gaussians, which is insufficient for learning a continuous SDF. Then, to address this limitation, we incorporate volumetric rendering and align the rendered geometric attributes (depth, normal) with that derived from 3DGS. In sum, these two designs allow SDF and 3DGS to be aligned, jointly optimized, and mutually boosted. Our extensive experimental results demonstrate that our 3DGSR enables high-quality 3D surface reconstruction while preserving the efficiency and rendering quality of 3DGS. Besides, our method competes favorably with leading surface reconstruction techniques while offering a more efficient learning process and much better rendering qualities.","PeriodicalId":50913,"journal":{"name":"ACM Transactions on Graphics","volume":"176 1","pages":""},"PeriodicalIF":6.2,"publicationDate":"2024-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142672827","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}
Kehan Xu, Sebastian Herholz, Marco Manzi, Marios Papas, Markus Gross
Simulating the light transport of volumetric effects poses significant challenges and costs, especially in the presence of heterogeneous volumes. Generating stochastic paths for volume rendering involves multiple decisions, and previous works mainly focused on directional and distance sampling, where the volume scattering probability (VSP), i.e., the probability of scattering inside a volume, is indirectly determined as a byproduct of distance sampling. We demonstrate that direct control over the VSP can significantly improve efficiency and present an unbiased volume rendering algorithm based on an existing resampling framework for precise control over the VSP. Compared to previous state-of-the-art, which can only increase the VSP without guaranteeing to reach the desired value, our method also supports decreasing the VSP. We further present a data-driven guiding framework to efficiently learn and query an approximation of the optimal VSP everywhere in the scene without the need for user control. Our approach can easily be combined with existing path-guiding methods for directional sampling at minimal overhead and shows significant improvements over the state-of-the-art in various complex volumetric lighting scenarios.
{"title":"Volume Scattering Probability Guiding","authors":"Kehan Xu, Sebastian Herholz, Marco Manzi, Marios Papas, Markus Gross","doi":"10.1145/3687982","DOIUrl":"https://doi.org/10.1145/3687982","url":null,"abstract":"Simulating the light transport of volumetric effects poses significant challenges and costs, especially in the presence of heterogeneous volumes. Generating stochastic paths for volume rendering involves multiple decisions, and previous works mainly focused on directional and distance sampling, where the volume scattering probability (VSP), i.e., the probability of scattering inside a volume, is indirectly determined as a byproduct of distance sampling. We demonstrate that direct control over the VSP can significantly improve efficiency and present an unbiased volume rendering algorithm based on an existing resampling framework for precise control over the VSP. Compared to previous state-of-the-art, which can only increase the VSP without guaranteeing to reach the desired value, our method also supports decreasing the VSP. We further present a data-driven guiding framework to efficiently learn and query an approximation of the optimal VSP everywhere in the scene without the need for user control. Our approach can easily be combined with existing path-guiding methods for directional sampling at minimal overhead and shows significant improvements over the state-of-the-art in various complex volumetric lighting scenarios.","PeriodicalId":50913,"journal":{"name":"ACM Transactions on Graphics","volume":"14 1","pages":""},"PeriodicalIF":6.2,"publicationDate":"2024-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142673045","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}
Learned optics, which incorporate lightweight diffractive optics, coded-aperture modulation, and specialized image-processing neural networks, have recently garnered attention in the field of snapshot hyperspectral imaging (HSI). While conventional methods typically rely on a single lens element paired with an off-the-shelf color sensor, these setups, despite their widespread availability, present inherent limitations. First, the Bayer sensor's spectral response curves are not optimized for HSI applications, limiting spectral fidelity of the reconstruction. Second, single lens designs rely on a single diffractive optical element (DOE) to simultaneously encode spectral information and maintain spatial resolution across all wavelengths, which constrains spectral encoding capabilities. This work investigates a multi-channel lens array combined with aperture-wise color filters, all co-optimized alongside an image reconstruction network. This configuration enables independent spatial encoding and spectral response for each channel, improving optical encoding across both spatial and spectral dimensions. Specifically, we validate that the method achieves over a 5dB improvement in PSNR for spectral reconstruction compared to existing single-diffractive lens and coded-aperture techniques. Experimental validation further confirmed that the method is capable of recovering up to 31 spectral bands within the 429--700 nm range in diverse indoor and outdoor environments.
{"title":"Learned Multi-aperture Color-coded Optics for Snapshot Hyperspectral Imaging","authors":"Zheng Shi, Xiong Dun, Haoyu Wei, Siyu Dong, Zhanshan Wang, Xinbin Cheng, Felix Heide, Yifan Peng","doi":"10.1145/3687976","DOIUrl":"https://doi.org/10.1145/3687976","url":null,"abstract":"Learned optics, which incorporate lightweight diffractive optics, coded-aperture modulation, and specialized image-processing neural networks, have recently garnered attention in the field of snapshot hyperspectral imaging (HSI). While conventional methods typically rely on a single lens element paired with an off-the-shelf color sensor, these setups, despite their widespread availability, present inherent limitations. First, the Bayer sensor's spectral response curves are not optimized for HSI applications, limiting spectral fidelity of the reconstruction. Second, single lens designs rely on a single diffractive optical element (DOE) to simultaneously encode spectral information and maintain spatial resolution across all wavelengths, which constrains spectral encoding capabilities. This work investigates a multi-channel lens array combined with aperture-wise color filters, all co-optimized alongside an image reconstruction network. This configuration enables independent spatial encoding and spectral response for each channel, improving optical encoding across both spatial and spectral dimensions. Specifically, we validate that the method achieves over a 5dB improvement in PSNR for spectral reconstruction compared to existing single-diffractive lens and coded-aperture techniques. Experimental validation further confirmed that the method is capable of recovering up to 31 spectral bands within the 429--700 nm range in diverse indoor and outdoor environments.","PeriodicalId":50913,"journal":{"name":"ACM Transactions on Graphics","volume":"25 1","pages":""},"PeriodicalIF":6.2,"publicationDate":"2024-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142673094","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}
Steffen Hinderink, Hendrik Brückler, Marcel Campen
A method for the construction of bijective volumetric maps between 3D shapes is presented. Arbitrary shapes of ball-topology are supported, overcoming restrictions of previous methods to convex or star-shaped targets. In essence, the mapping problem is decomposed into a set of simpler mapping problems, each of which can be solved with previous methods for discrete star-shaped mapping problems. Addressing the key challenges in this endeavor, algorithms are described to reliably construct structurally compatible partitions of two shapes with constraints regarding star-shapedness and to compute a parsimonious common refinement of two triangulations.
{"title":"Bijective Volumetric Mapping via Star Decomposition","authors":"Steffen Hinderink, Hendrik Brückler, Marcel Campen","doi":"10.1145/3687950","DOIUrl":"https://doi.org/10.1145/3687950","url":null,"abstract":"A method for the construction of bijective volumetric maps between 3D shapes is presented. Arbitrary shapes of ball-topology are supported, overcoming restrictions of previous methods to convex or star-shaped targets. In essence, the mapping problem is decomposed into a set of simpler mapping problems, each of which can be solved with previous methods for discrete star-shaped mapping problems. Addressing the key challenges in this endeavor, algorithms are described to reliably construct structurally compatible partitions of two shapes with constraints regarding star-shapedness and to compute a parsimonious common refinement of two triangulations.","PeriodicalId":50913,"journal":{"name":"ACM Transactions on Graphics","volume":"25 1","pages":""},"PeriodicalIF":6.2,"publicationDate":"2024-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142673119","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}
We propose a method to reproduce dynamic appearance textures with space-stationary but time-varying visual statistics. While most previous work decomposes dynamic textures into static appearance and motion, we focus on dynamic appearance that results not from motion but variations of fundamental properties, such as rusting, decaying, melting, and weathering. To this end, we adopt the neural ordinary differential equation (ODE) to learn the underlying dynamics of appearance from a target exemplar. We simulate the ODE in two phases. At the "warm-up" phase, the ODE diffuses a random noise to an initial state. We then constrain the further evolution of this ODE to replicate the evolution of visual feature statistics in the exemplar during the generation phase. The particular innovation of this work is the neural ODE achieving both denoising and evolution for dynamics synthesis, with a proposed temporal training scheme. We study both relightable (BRDF) and non-relightable (RGB) appearance models. For both we introduce new pilot datasets, allowing, for the first time, to study such phenomena: For RGB we provide 22 dynamic textures acquired from free online sources; For BRDFs, we further acquire a dataset of 21 flash-lit videos of time-varying materials, enabled by a simple-to-construct setup. Our experiments show that our method consistently yields realistic and coherent results, whereas prior works falter under pronounced temporal appearance variations. A user study confirms our approach is preferred to previous work for such exemplars.
{"title":"Neural Differential Appearance Equations","authors":"Chen Liu, Tobias Ritschel","doi":"10.1145/3687900","DOIUrl":"https://doi.org/10.1145/3687900","url":null,"abstract":"We propose a method to reproduce dynamic appearance textures with space-stationary but time-varying visual statistics. While most previous work decomposes dynamic textures into static appearance and motion, we focus on dynamic appearance that results not from motion but variations of fundamental properties, such as rusting, decaying, melting, and weathering. To this end, we adopt the neural ordinary differential equation (ODE) to learn the underlying dynamics of appearance from a target exemplar. We simulate the ODE in two phases. At the \"warm-up\" phase, the ODE diffuses a random noise to an initial state. We then constrain the further evolution of this ODE to replicate the evolution of visual feature statistics in the exemplar during the generation phase. The particular innovation of this work is the neural ODE achieving both denoising and evolution for dynamics synthesis, with a proposed temporal training scheme. We study both relightable (BRDF) and non-relightable (RGB) appearance models. For both we introduce new pilot datasets, allowing, for the first time, to study such phenomena: For RGB we provide 22 dynamic textures acquired from free online sources; For BRDFs, we further acquire a dataset of 21 flash-lit videos of time-varying materials, enabled by a simple-to-construct setup. Our experiments show that our method consistently yields realistic and coherent results, whereas prior works falter under pronounced temporal appearance variations. A user study confirms our approach is preferred to previous work for such exemplars.","PeriodicalId":50913,"journal":{"name":"ACM Transactions on Graphics","volume":"53 1","pages":""},"PeriodicalIF":6.2,"publicationDate":"2024-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142673122","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}
Bailey Miller, Rohan Sawhney, Keenan Crane, Ioannis Gkioulekas
We introduce a Monte Carlo method for computing derivatives of the solution to a partial differential equation (PDE) with respect to problem parameters (such as domain geometry or boundary conditions). Derivatives can be evaluated at arbitrary points, without performing a global solve or constructing a volumetric grid or mesh. The method is hence well suited to inverse problems with complex geometry, such as PDE-constrained shape optimization. Like other walk on spheres (WoS) algorithms, our method is trivial to parallelize, and is agnostic to boundary representation (meshes, splines, implicit surfaces, etc. ), supporting large topological changes. We focus in particular on screened Poisson equations, which model diverse problems from scientific and geometric computing. As in differentiable rendering, we jointly estimate derivatives with respect to all parameters---hence, cost does not grow significantly with parameter count. In practice, even noisy derivative estimates exhibit fast, stable convergence for stochastic gradient-based optimization, as we show through examples from thermal design, shape from diffusion, and computer graphics.
{"title":"Differential Walk on Spheres","authors":"Bailey Miller, Rohan Sawhney, Keenan Crane, Ioannis Gkioulekas","doi":"10.1145/3687913","DOIUrl":"https://doi.org/10.1145/3687913","url":null,"abstract":"We introduce a Monte Carlo method for computing derivatives of the solution to a partial differential equation (PDE) with respect to problem parameters (such as domain geometry or boundary conditions). Derivatives can be evaluated at arbitrary points, without performing a global solve or constructing a volumetric grid or mesh. The method is hence well suited to inverse problems with complex geometry, such as PDE-constrained shape optimization. Like other <jats:italic>walk on spheres (WoS)</jats:italic> algorithms, our method is trivial to parallelize, and is agnostic to boundary representation (meshes, splines, implicit surfaces, <jats:italic>etc.</jats:italic> ), supporting large topological changes. We focus in particular on screened Poisson equations, which model diverse problems from scientific and geometric computing. As in differentiable rendering, we jointly estimate derivatives with respect to all parameters---hence, cost does not grow significantly with parameter count. In practice, even noisy derivative estimates exhibit fast, stable convergence for stochastic gradient-based optimization, as we show through examples from thermal design, shape from diffusion, and computer graphics.","PeriodicalId":50913,"journal":{"name":"ACM Transactions on Graphics","volume":"39 1","pages":""},"PeriodicalIF":6.2,"publicationDate":"2024-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142672833","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}