Filters








10,521 Hits in 2.4 sec

Overfitting Control for Surface Reconstruction [article]

Yunjin Lee, Seungyong Lee, Ioannis Ivrissimtzis, Hans-Peter Seidel
2006 Symposium on geometry processing : [proceedings]. Symposium on Geometry Processing  
This paper proposes a general framework for overfitting control in surface reconstruction from noisy point data.  ...  It is fully automatic and can work in conjunction with any surface reconstruction algorithm.  ...  Discussion and Future Work We propose a framework for the systematic control of overfitting in surface reconstruction.  ... 
doi:10.2312/sgp/sgp06/231-234 fatcat:seqqzjiebrhwbhgsziiiq65xkm

Self-Organising Maps for Implicit Surface Reconstruction [article]

Mincheol Yoon, Ioannis Ivrissimtzis, Seungyong Lee
2008 Computer Graphics and Visual Computing  
This paper proposes an implicit surface reconstruction algorithm based on Self-Organising Maps (SOMs).  ...  After each training iteration, we use extra sample validation to test for overfitting. At the end of the training process, a triangle mesh is extracted as the zero level set of the SOM grid.  ...  See Fig. 1 for an example of overfitting in curve reconstruction. In this paper we use extra sample validation to control overfitting.  ... 
doi:10.2312/localchapterevents/tpcg/tpcg08/083-090 dblp:conf/tpcg/YoonIL08 fatcat:7z43cnqjz5gatgo55ssv3ezwwq

Extending Neural Networks for B-Spline Surface Reconstruction [chapter]

G. Echevarría, A. Iglesias, A. Gálvez
2002 Lecture Notes in Computer Science  
In this paper, the use of B-spline surfaces (by far, the most common family of surfaces in surface modeling and industry) for the surface reconstruction problem is proposed instead.  ...  This analysis also includes the use of two sets of data (the training and the testing data) to check for overfitting, which does not occur here.  ...  Table 3 shows the control points for the reconstructed Surface I corresponding to the best case M = N = 6.  ... 
doi:10.1007/3-540-46080-2_32 fatcat:skcjl7xurbggrh7hho53qhxiay

DeepCurrents: Learning Implicit Representations of Shapes with Boundaries [article]

David Palmer and Dmitriy Smirnov and Stephanie Wang and Albert Chern and Justin Solomon
2022 arXiv   pre-print
Many of these methods, however, learn only closed surfaces and are unable to reconstruct shapes with boundary curves.  ...  Recent techniques have been successful in reconstructing surfaces as level sets of learned functions (such as signed distance fields) parameterized by deep neural networks.  ...  Human body surface reconstructions. We overfit Deep-Currents models to reconstruct several torso, head, hand, and foot meshes.  ... 
arXiv:2111.09383v2 fatcat:l7jv62o5jrevddktahtmph5n5e

Iso-Points: Optimizing Neural Implicit Surfaces with Hybrid Representations [article]

Wang Yifan, Shihao Wu, Cengiz Oztireli, Olga Sorkine-Hornung
2021 arXiv   pre-print
We demonstrate that our method can be adopted to improve state-of-the-art techniques for reconstructing neural implicit surfaces from multi-view images or point clouds.  ...  Neural implicit functions have emerged as a powerful representation for surfaces in 3D.  ...  Acknowledgement We would like to thank Viviane Yang for her help with the point2surf code.  ... 
arXiv:2012.06434v2 fatcat:guysfuk2b5cnjpljmb4hslezyu

Free energy surface reconstruction from umbrella samples using Gaussian process regression II: Multiple collective variables [article]

Noam Bernstein, Thomas Stecher, Gábor Csányi
2014 arXiv   pre-print
In Part I we review the necessary theory and test the method for one collective variable.  ...  In Part II we consider the case of multiple collective variables and compare to a reconstruction using least squares fitting of radial basis functions.  ...  Acknowledgement We thank Eric Vanden-Eijnden for comments on the manuscript. N.B. acknowledges fund-  ... 
arXiv:1312.4420v2 fatcat:vdobkmf3ybgdff7hsblukvgzwm

Local Geometric Indexing of High Resolution Data for Facial Reconstruction from Sparse Markers [article]

Matthew Cong, Lana Lan, Ronald Fedkiw
2019 arXiv   pre-print
When considering sparse motion capture marker data, one typically struggles to balance its overfitting via a high dimensional blendshape system versus underfitting caused by smoothness constraints.  ...  With the current trend towards using more and more data, our aim is not to fit the motion capture markers with a parameterized (blendshape) model or to smoothly interpolate a surface through the marker  ...  Acknowledgements We would like to thank Cary Phillips, Brian Cantwell, Kevin Sprout, and Industrial Light & Magic for supporting our research into facial performance capture.  ... 
arXiv:1903.00119v2 fatcat:6be527nrgfdgnfzt2c22okyt54

Locally Developable Constraint for Document Surface Reconstruction

Yuanlong Shao, Xinguo Liu, Xueying Qin, Yi Xu, Hujun Bao
2009 2009 10th International Conference on Document Analysis and Recognition  
This article presents a global optimization approach to reconstruct surfaces from a single document image.  ...  We show that even this is enough to reconstruct a wide class of document surfaces even with an uncalibrated camera.  ...  All these provide structural information for document surface reconstruction.  ... 
doi:10.1109/icdar.2009.57 dblp:conf/icdar/ShaoLQXB09 fatcat:juexdgxdhvappevzcnixfna62y

Robust flow field reconstruction from limited measurements via sparse representation [article]

Jared Callaham and Kazuki Maeda and Steven L. Brunton
2019 arXiv   pre-print
Sparse representation has been widely used for image recognition and reconstruction, and it is well-suited to structured data with limited, corrupt measurements.  ...  We explore sparse representation for flow reconstruction on a variety of fluid data sets with a wide range of complexity, including vortex shedding past a cylinder at low Reynolds number, a mixing layer  ...  Special thanks to Nathan Kutz for sharing his insights into the extreme utility of sparsity promoting methods for complex physical systems.  ... 
arXiv:1810.06723v2 fatcat:eq7gdbtjnbh3xkwa2kemdbgmme

Atomic Image Reconstruction from Atomic Resolution Holography Using L1-Regularized Linear Regression

Tomohiro Matsushita
2016 e-Journal of Surface Science and Nanotechnology  
The three-dimensional atomic image is reconstructed from the inverse x-ray fluorescence hologram using a calculation. The performance of the algorithm used for this calculation is important.  ...  However, images reconstructed by this method sometimes contain artifacts. I developed an algorithm (SPEA-L1) using an iterative method with L1 regularization and report its performance.  ...  The 3D atomic image is reconstructed from the hologram using a calculation. The performance of the algorithm used for the reconstruction calculation is quite important.  ... 
doi:10.1380/ejssnt.2016.158 fatcat:2q3u54f5ufce7m4ljw7lnavu2a

Neural Geometric Level of Detail: Real-time Rendering with Implicit 3D Shapes [article]

Towaki Takikawa, Joey Litalien, Kangxue Yin, Karsten Kreis, Charles Loop, Derek Nowrouzezahrai, Alec Jacobson, Morgan McGuire, Sanja Fidler
2021 arXiv   pre-print
Furthermore, it produces state-of-the-art reconstruction quality for complex shapes under both 3D geometric and 2D image-space metrics.  ...  We introduce an efficient neural representation that, for the first time, enables real-time rendering of high-fidelity neural SDFs, while achieving state-of-the-art geometry reconstruction quality.  ...  We also thank Jacob Munkberg, Peter Shirley, Alexander Majercik, David Luebke, Jonah Philion, and Jun Gao for help with paper review.  ... 
arXiv:2101.10994v1 fatcat:w3sgfbn4kzahtcr2uwnlaivgey

AdaFit: Rethinking Learning-based Normal Estimation on Point Clouds [article]

Runsong Zhu, Yuan Liu, Zhen Dong, Tengping Jiang, Yuan Wang, Wenping Wang, Bisheng Yang
2021 arXiv   pre-print
Existing works use a network to learn point-wise weights for weighted least squares surface fitting to estimate the normals, which has difficulty in finding accurate normals in complex regions or containing  ...  By analyzing the step of weighted least squares surface fitting, we find that it is hard to determine the polynomial order of the fitting surface and the fitting surface is sensitive to outliers.  ...  to control the contribution of every neighboring point to the final surface.  ... 
arXiv:2108.05836v1 fatcat:bhnqdg4aizgrthwgdv4gvqjk64

An improved empirical dynamic control system model of global mean sea level rise and surface temperature change

Qing Wu, Quang-Hung Luu, Pavel Tkalich, Ge Chen
2017 Theoretical and Applied Climatology  
We propose an improved statistical reconstruction model based on the empirical dynamic control system by taking into account the climate variability and deriving parameters from Monte Carlo cross-validation  ...  The averaged root mean square errors are reduced in both reconstructed fields, namely, the global mean surface temperature (by 24-37%) and the global mean sea level (by 5-25%).  ...  In fact, archiving the reconstructed parameters though applying a single least squares estimate could potentially contain the risk of overfitting, which will not be suitable for extending the method to  ... 
doi:10.1007/s00704-017-2039-3 fatcat:sgub5dm2cng4fj2sdl7eyo4xey

Functional networks for B-spline surface reconstruction

A. Iglesias, G. Echevarría, A. Gálvez
2004 Future generations computer systems  
In this paper, the use of B-spline surfaces (by far the most common family of surfaces in surface modeling and industry) for the surface reconstruction problem is proposed instead.  ...  This analysis also includes the use of two sets of data (the training and the testing data) to check for overfitting, which does not occur here.  ...  Table 3 shows the control points for the reconstructed Surface I corresponding to the best case M = N = 6.  ... 
doi:10.1016/j.future.2004.05.025 fatcat:4k4ljfwxjvf63gf4hkq2hnunu4

Reconstruction of basal properties in ice sheets using iterative inverse methods

Marijke Habermann, David Maxwell, Martin Truffer
2012 Journal of Glaciology  
derived surface velocities as input data.  ...  Here, iterative inverse methods are implemented for the specific problem of reconstructing basal stickiness of an ice sheet by using the shallow-shelf approximation as a forward model and synthetically  ...  Ian Joughin provided data for the funnelshaped ice stream tests. Ed Bueler, Constantine Khroulev, Andy Aschwanden and Ronni Grapenthin have patiently answered questions and provided feedback.  ... 
doi:10.3189/2012jog11j168 fatcat:izdl57754nezjdxotzvkdzg44u
« Previous Showing results 1 — 15 out of 10,521 results