Filters








2,272 Hits in 4.6 sec

Generating Liquid Simulations with Deformation-aware Neural Networks [article]

Lukas Prantl, Boris Bonev, Nils Thuerey
2019 arXiv   pre-print
We propose a novel approach for deformation-aware neural networks that learn the weighting and synthesis of dense volumetric deformation fields.  ...  Our algorithm captures these complex phenomena in two stages: a first neural network computes a weighting function for a set of pre-computed deformations, while a second network directly generates a deformation  ...  Appendix: Generating Liquid Simulations with Deformation-aware Neural Networks This supplemental document will first detail the necessary steps to align multiple, weighted deformation fields.  ... 
arXiv:1704.07854v4 fatcat:benfmgsv6rf4lclrvceesp4py4

Neural UpFlow: A Scene Flow Learning Approach to Increase the Apparent Resolution of Particle-Based Liquids [article]

Bruno Roy, Pierre Poulin, Eric Paquette
2021 arXiv   pre-print
We present a novel up-resing technique for generating high-resolution liquids based on scene flow estimation using deep neural networks.  ...  We furthermore showcase how the proposed interpolation approach can facilitate generating large simulation datasets with a subset of initial condition parameters.  ...  They propose to leverage neural networks to learn precomputed deformations of liquid surfaces.  ... 
arXiv:2106.05143v1 fatcat:yzb76xpbrrbstnyi6c7nwqnvby

Deformation-Aware Data-Driven Grasp Synthesis [article]

Tran Nguyen Le, Jens Lundell, Fares J.Abu-Dakka, Ville Kyrki
2021 arXiv   pre-print
We also curate a new synthetic dataset of grasps on objects of varying stiffness using the Isaac Gym simulator for training the network.  ...  However, deformations are not necessarily harmful -- humans are, for example, able to exploit deformations to generate new potential grasps. How to achieve that on a robot is though an open question.  ...  by a deep neural network trained on purely synthetic data.  ... 
arXiv:2109.05320v1 fatcat:r7lljmp4praxvpzxdxlsv5pv2q

Realtime Simulation of Thin-Shell Deformable Materials using CNN-Based Mesh Embedding [article]

Qingyang Tan, Zherong Pan, Lin Gao, Dinesh Manocha
2020 arXiv   pre-print
Our key technique is a graph-based convolutional neural network (CNN) defined on meshes with arbitrary topologies and a new mesh embedding approach based on physics-inspired loss term.  ...  Finally, we show that the temporal evolution of the mesh in the feature space can also be learned using a recurrent neural network (RNN) leading to fully learnable physics simulators.  ...  To make the method aware of physics rules, we augment the embedding network with a stateful feature space simulator represented as a MLP.  ... 
arXiv:1909.12354v4 fatcat:ubmxxpbii5antlvv4rbv2ujype

Data-driven synthesis of smoke flows with CNN-based feature descriptors

Mengyu Chu, Nils Thuerey
2017 ACM Transactions on Graphics  
We use convolutional neural networks to generate the descriptors from fluid data such as smoke density and flow velocity.  ...  With the help of this patch advection, we generate stable space-time data sets from detailed fluids for our repositories.  ...  ACKNOWLEDGMENTS The authors would like to thank Marie-Lena Eckert for helping with generating the video, and Daniel Hook for his work on the direct density synthesis with CNNs.  ... 
doi:10.1145/3072959.3073643 fatcat:4cp4hqmnyzavhndqrtzoe75xbq

2020 Index IEEE Transactions on Visualization and Computer Graphics Vol. 26

2021 IEEE Transactions on Visualization and Computer Graphics  
Wentzel, A., +, TVCG Jan. 2020 949-959 DeepOrganNet: On-the-Fly Reconstruction and Visualization of 3D / 4D Lung Models from Single-View Projections by Deep Deformation Network.  ...  Medical image processing CerebroVis: Designing an Abstract yet Spatially Contextualized Cerebral Artery Network Visualization.  ...  ., Simulating Liquids on Dynamically Warping Grids; TVCG June 2020 2288-2302 Igarashi, T., see Ibayashi, H., 2288-2302 Isaacs, K., see Williams, K., 1118-1128 Isenberg, P., see Chang, R., TVCG Jan. 2020  ... 
doi:10.1109/tvcg.2021.3111804 fatcat:lvhpoz5sqjhclo3roocrdtzs2m

Identification of Granule Growth Regimes in High Shear Wet Granulation Processes Using a Physics-Constrained Neural Network

Chaitanya Sampat, Rohit Ramachandran
2021 Processes  
The physics-constrained neural network (PCNN) was better at predicting granule growth regimes when compared to other neural networks with no physical constraints.  ...  When input data that violated physics-based boundaries was provided, the PCNN identified these points more accurately compared to other non-physics constrained neural networks, with an error of <1%.  ...  Acknowledgments: The authors would also like to acknowledge Anik Chaturbedi and Indu Muthancheri for their help with development of the PBM code and inputs.  ... 
doi:10.3390/pr9050737 doaj:cf82c714bcdc408ebd5b092e11c7debe fatcat:nt2bn7o4gnhhnpjuso7k6hemoe

Determination of penetration depth at high velocity impact using finite element method and artificial neural network tools

Namık KılıÇ, Bülent Ekici, Selim Hartomacıoğlu
2015 Defence Technology  
This study aims to apply a hybrid method using FEM simulation and artificial neural network (ANN) analysis to approximate ballistic limit thickness for armor steels.  ...  In this methodology, the FEM simulations are used to create training cases for Multilayer Perceptron (MLP) three layer networks.  ...  of a large data set, reasonable results can be achieved. 3) Training results itself not sufficient to give the decision on neural network topology. 4) The confidence of the data generated with FE simulations  ... 
doi:10.1016/j.dt.2014.12.001 fatcat:5m6vcwxhx5df3parjvdgbg4oem

2020 Index IEEE Transactions on Semiconductor Manufacturing Vol. 33

2020 IEEE transactions on semiconductor manufacturing  
Bhat, T.S., +, TSM May 2020 291-294 Generative adversarial networks Combination of Convolutional and Generative Adversarial Networks for Defect Image Demoiréing of Thin-Film Transistor Liquid-Crystal  ...  film transistors Combination of Convolutional and Generative Adversarial Networks for Defect Image Demoiréing of Thin-Film Transistor Liquid-Crystal Display Image.  ... 
doi:10.1109/tsm.2020.3036306 fatcat:mb4mmpjd4vbzfep47oxhm3soca

Intelligent Systems for Detection and Control Damage on Buried Infrastructure due to Soil Subsidence

Silvia García
2020 Civil Engineering Research Journal  
of the pumping of the aquifers, simulation of catastrophic scenarios for the reintegration of the inhabitants in sites with better capacities to hold communities.  ...  unbalanced consumption of the vital liquid.  ...  At these points a sinking process had already been registered that alerted the inhabitants and forced the emergency teams to deal with constant leaks in the liquid conduction pipes but during the earthquake  ... 
doi:10.19080/cerj.2019.09.555771 fatcat:kqcfo23j6vbxxo2onb2rfunxca

Subspace Graph Physics: Real-Time Rigid Body-Driven Granular Flow Simulation [article]

Amin Haeri, Krzysztof Skonieczny
2021 arXiv   pre-print
A graph network simulator (GNS) is trained to learn the underlying subspace dynamics. The learned GNS is then able to predict particle positions and interaction forces with good accuracy.  ...  An important challenge in robotics is understanding the interactions between robots and deformable terrains that consist of granular material.  ...  Liquid splash modeling with neural networks. Computer Graphics Forum 37, 171– 182.  ... 
arXiv:2111.10206v1 fatcat:vntwi2zoevfsrfmvohvydhqqte

Latent Space Subdivision: Stable and Controllable Time Predictions for Fluid Flow [article]

Steffen Wiewel, Byungsoo Kim, Vinicius C. Azevedo, Barbara Solenthaler, Nils Thuerey
2020 arXiv   pre-print
To achieve stable predictions for long-term flow sequences, a convolutional neural network (CNN) is trained for spatial compression in combination with a temporal prediction network that consists of stacked  ...  We propose an end-to-end trained neural networkarchitecture to robustly predict the complex dynamics of fluid flows with high temporal stability.  ...  Kim et al. (2019) introduced a generative deep neural network for parameterized fluid simulations that only takes a small set of physical parameters as input to very efficiently synthesize points in the  ... 
arXiv:2003.08723v1 fatcat:ovdf7njbhfh5lpeswa6olgubxm

Neuroevolutionary Multiobjective Methodology for the Optimization of the Injection Blow Molding Process [chapter]

Renê Pinto, Hugo Silva, Fernando Duarte, Joao Nunes, Antonio Gaspar-Cunha
2019 Lecture Notes in Computer Science  
This work has been supported by FCT -Fundação para a Ciência e Tecnologia in the scope of the project: PEst-OE/EEI/UI0319/2014 and the European project MSCA-RISE-2015, NEWEX, with reference 734205.  ...  Artificial Neural Networks (ANNs) has been used in several studies to describe blow molding process with high accuracy.  ...  After being generated, each individual is evaluated by a procedure that comprises assembly the neural network from chromosome information and fed into the network the coordinates of each point of the finite  ... 
doi:10.1007/978-3-030-12598-1_59 fatcat:5t7ynivrlbgordq3hovvl6aqiq

Transport-Based Neural Style Transfer for Smoke Simulations [article]

Byungsoo Kim, Vinicius C. Azevedo, Markus Gross, Barbara Solenthaler
2019 arXiv   pre-print
Our method is able to transfer features from natural images to smoke simulations, enabling general content-aware manipulations ranging from simple patterns to intricate motifs.  ...  In this paper, we propose the first transport-based Neural Style Transfer (TNST) algorithm for volumetric smoke data.  ...  Our method is able to transfer features from natural images to smoke simulations, enabling general content-aware manipulations ranging from simple patterns to intricate motifs.  ... 
arXiv:1905.07442v1 fatcat:ryaghfq5wjhj3fyyt7jthgwz6q

Study of the Permanent Deformation of Soil Used in Flexible Pavement Design

Wendel S. Cabral, Suelly H. A. Barroso, Samuel A. Torquato, Qiang Tang
2020 Advances in Civil Engineering  
This is a study of the permanent deformation (PD) of soil used in pavement layers, obtaining prediction models through the technique of artificial neural networks, in addition to the design of pavement  ...  The decreasing supply of soils with geotechnical parameters suitable for pavement designs is a visible problem in our environment.  ...  After several attempts and analyses to obtain the best explanatory topology of the neural model and being aware that there are other possible combinations to reach the most satisfactory network, it was  ... 
doi:10.1155/2020/4274926 fatcat:2ezzkxvrqjbsvlpdux66mfjc2y
« Previous Showing results 1 — 15 out of 2,272 results