In this paper we provide, first, a general symbolic algorithm for computing the symmetries of a given rational surface, based on the classical differential invariants of surfaces, i.e. Gauss curvature and mean curvatu...
详细信息
Image stitching technology has application scenarios in many fields. At present, the existing algorithms usually use the method of acquisition first and then synthesis, which is still insufficient in real-time perform...
详细信息
ISBN:
(数字)9798350388725
ISBN:
(纸本)9798350388732
Image stitching technology has application scenarios in many fields. At present, the existing algorithms usually use the method of acquisition first and then synthesis, which is still insufficient in real-time performance; in addition, under the premise of real-time acquisition, it is also insufficient to realize continuous image synthesis. To solve this problem, this algorithm uses the RGB module of the Intel Realsense D435 camera for image acquisition, first creates an original image for storing the final result, collects an image at an interval of 100ms each time, and performs 8*8 hashing of the two images before and after Compare the similarity of the value, take the image with a similarity less than 5/8 and keep it, then use the SIFT (Scale-invariant feature transform) scale-invariant feature transform feature detection algorithm to extract the image feature points from the collected image, and use the RANSAC to extract the feature points The (Random Sample consensus) algorithm screens effective points and calculates the homography transformation matrix at the same time. Finally, every two images collected are synthesized into one and covered in the corresponding position of the result image. Through experiments in this paper, the average time for single acquisition and synthesis is 70ms, achieving the real-time goal. The similarity between the experimental group and the control group can reach 70%, and the resolution is increased by 1.65 times, achieving the goal of continuous splicing.
This paper focuses on a class of inclusion problems of maximal monotone operators over directed multi-agent networks, where each agent is characterized by a splittable local operator. By integrating the ideas from cor...
详细信息
Empirical risk minimization, where the underlying loss function depends on a pair of data points, covers a wide range of application areas in statistics including pairwise ranking and survival analysis. The common emp...
详细信息
Empirical risk minimization, where the underlying loss function depends on a pair of data points, covers a wide range of application areas in statistics including pairwise ranking and survival analysis. The common empirical risk estimator obtained by averaging values of a loss function over all possible pairs of observations is essentially a U-statistic. One well-known problem with minimizing U-statistic type empirical risks, is that the computa-tional complexity of U-statistics increases quadratically with the sample size. When faced with big data, this poses computational challenges as the colossal number of observation pairs virtually prohibits centralized computing to be performed on a single machine. This paper addresses this problem by developing two computationally and statistically efficient methods based on the divide-and-conquer strategy on a decentralized computing system, whereby the data are distributed among machines to perform the tasks. One of these methods is based on a surrogate of the empirical risk, while the other method extends the one-step updating scheme in classical M-estimation to the case of pairwise loss. We show that the proposed estimators are as asymptotically efficient as the benchmark global U-estimator obtained under centralized computing. As well, we introduce two distributed iterative algorithms to facilitate the implementation of the proposed methods, and conduct extensive numerical experiments to demonstrate their merit.
This paper proposes a two-timescale compressed primal-dual (TiCoPD) algorithm for decentralized optimization with improved communication efficiency over prior works on primal-dual decentralized optimization. The algor...
详细信息
We consider the minimization of an M-convex function, which is a discrete convexity concept for functions on the integer lattice points. It is known that a minimizer of an M-convex function can be obtained by the stee...
详细信息
In this paper, we consider L2-approximation of functions in a weighted Korobov space. We consider a median algorithm, which is related to median integration rules, that have recently gained a lot of interest in the th...
详细信息
The Rashomon set of equally-good models promises less discriminatory algorithms, reduced outcome homogenization, and fairer decisions through model ensembles or reconciliation. However, we argue from the perspective o...
详细信息
We incorporate an iteratively reweighted strategy in the manifold proximal point algorithm (ManPPA) in [12] to solve an enhanced sparsity inducing model for identifying sparse yet nonzero vectors in a given subspace. ...
详细信息
We give the first local algorithm for computing multi-commodity flow and apply it to obtain a (1+∊)-approximate algorithm for computing a k-commodity flow on an expander with m edges in (m + ∊−3k3D)no(1) time, where D...
详细信息
暂无评论