اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدHLO: Half-kernel Laplacian Operator for Surface Smoothing
71
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
This paper presents a simple yet effective method for feature-preserving surface smoothing. Through analyzing the differential property of surfaces, we show that the conventional discrete Laplacian operator with uniform weights is not applicable to feature points at which the surface is non-differentiable and the second order derivatives do not exist. To overcome this difficulty, we propose a Half-kernel Laplacian Operator (HLO) as an alternative to the conventional Laplacian. Given a vertex v, HLO first finds all pairs of its neighboring vertices and divides each pair into two subsets (called half windows); then computes the uniform Laplacians of all such subsets and subsequently projects the computed Laplacians to the full-window uniform Laplacian to alleviate flipping and degeneration. The half window with least regularization energy is then chosen for v. We develop an iterative approach to apply HLO for surface denoising. Our method is conceptually simple and easy to use because it has a single parameter, i.e., the number of iterations for updating vertices. We show that our method can preserve features better than the popular uniform Laplacian-based denoising and it significantly alleviates the shrinkage artifact. Extensive experimental results demonstrate that HLO is better than or comparable to state-of-the-art techniques both qualitatively and quantitatively and that it is particularly good at handling meshes with high noise. We will make our source code publicly available.
قيم البحث
اقرأ أيضاً
We show that the driving force behind the regularizing effect of Laplacian smoothing on surface elements is the popular mean ratio quality measure. We use these insights to provide natural generalizations to polygons and polyhedra. The corresponding
functions measuring the quality of meshes are easily seen to be convex and can be used for global optimization-based untangling and smoothing. Using a simple backtracking line-search we compare several smoothing methods with respect to the resulting mesh quality. We also discuss their effectiveness in combination with topology modification.
In this paper, we investigate the efficiency of various strategies for subdividing polynomial triangular surface patches. We give a simple algorithm performing a regular subdivision in four calls to the standard de Casteljau algorithm (in its subdivi
sion version). A naive version uses twelve calls. We also show that any method for obtaining a regular subdivision using the standard de Casteljau algorithm requires at least 4 calls. Thus, our method is optimal. We give another subdivision algorithm using only three calls to the de Casteljau algorithm. Instead of being regular, the subdivision pattern is diamond-like. Finally, we present a ``spider-like subdivision scheme producing six subtriangles in four calls to the de Casteljau algorithm.
High dimensional B-splines are catching tremendous attentions in fields of Iso-geometry Analysis, dynamic surface reconstruction and so on. However, the actual measured data are usually sparse and nonuniform, which might not meet the requirement of t
raditional B-spline algorithms. In this paper, we present a novel dynamic surface reconstruction approach, which is a 3-dimensional key points interpolation method (KPI) based on B-spline, aimed at dealing with sparse distributed data. This method includes two stages: a data set generation algorithm based on Kriging and a control point solving method based on key points interpolation. The data set generation method is designed to construct a grided dataset which can meet the requirement of B-spline interpolation, while promisingly catching the trend of sparse data, and it also includes a parameter reduction method which can significantly reduce the number of control points of the result surface. The control points solving method ensures the 3-dimensional B-spline function to interpolate the sparse data points precisely while approximating the data points generated by Kriging. We apply the method into a temperature data interpolation problem. It is shown that the generated dynamic surface accurately interpolates the sparsely distributed temperature data, preserves the dynamic characteristics, with fewer control points than those of traditional B-spline surface interpolation algorithm.
We propose the adversarially robust kernel smoothing (ARKS) algorithm, combining kernel smoothing, robust optimization, and adversarial training for robust learning. Our methods are motivated by the convex analysis perspective of distributionally rob
ust optimization based on probability metrics, such as the Wasserstein distance and the maximum mean discrepancy. We adapt the integral operator using supremal convolution in convex analysis to form a novel function majorant used for enforcing robustness. Our method is simple in form and applies to general loss functions and machine learning models. Furthermore, we report experiments with general machine learning models, such as deep neural networks, to demonstrate that ARKS performs competitively with the state-of-the-art methods based on the Wasserstein distance.
Federated learning aims to protect data privacy by collaboratively learning a model without sharing private data among users. However, an adversary may still be able to infer the private training data by attacking the released model. Differential pri
vacy provides a statistical protection against such attacks at the price of significantly degrading the accuracy or utility of the trained models. In this paper, we investigate a utility enhancement scheme based on Laplacian smoothing for differentially private federated learning (DP-Fed-LS), where the parameter aggregation with injected Gaussian noise is improved in statistical precision without losing privacy budget. Our key observation is that the aggregated gradients in federated learning often enjoy a type of smoothness, i.e. sparsity in the graph Fourier basis with polynomial decays of Fourier coefficients as frequency grows, which can be exploited by the Laplacian smoothing efficiently. Under a prescribed differential privacy budget, convergence error bounds with tight rates are provided for DP-Fed-LS with uniform subsampling of heterogeneous Non-IID data, revealing possible utility improvement of Laplacian smoothing in effective dimensionality and variance reduction, among others. Experiments over MNIST, SVHN, and Shakespeare datasets show that the proposed method can improve model accuracy with DP-guarantee and membership privacy under both uniform and Poisson subsampling mechanisms.
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
