Hasil untuk "Oratory. Elocution, etc."

Varun Burde, Artem Moroz, Vit Zeman et al.

Object pose estimation is a prominent task in computer vision. The object pose gives the orientation and translation of the object in real-world space, which allows various applications such as manipulation, augmented reality, etc. Various objects exhibit different properties with light, such as reflections, absorption, etc. This makes it challenging to understand the object's structure in RGB and depth channels. Recent research has been moving toward learning-based methods, which provide a more flexible and generalizable approach to object pose estimation utilizing deep learning. One such approach is the render-and-compare method, which renders the object from multiple views and compares it against the given 2D image, which often requires an object representation in the form of a CAD model. We reason that the synthetic texture of the CAD model may not be ideal for rendering and comparing operations. We showed that if the object is represented as an implicit (neural) representation in the form of Neural Radiance Field (NeRF), it exhibits a more realistic rendering of the actual scene and retains the crucial spatial features, which makes the comparison more versatile. We evaluated our NeRF implementation of the render-and-compare method on transparent datasets and found that it surpassed the current state-of-the-art results.

Prabhat Kumar Chand, Anisur Rahaman Molla, Sumathi Sivasubramaniam

Research involving computing with mobile agents is a fast-growing field, given the advancement of technology in automated systems, e.g., robots, drones, self-driving cars, etc. Therefore, it is pressing to focus on solving classical network problems using mobile agents. In this paper, we study one such problem -- finding small dominating sets of a graph $G$ using mobile agents. Dominating set is interesting in the field of mobile agents as it opens up a way for solving various robotic problems, e.g., guarding, covering, facility location, transport routing, etc. In this paper, we first present two algorithms for computing a {\em minimal dominating set}: (i) an $O(m)$ time algorithm if the robots start from a single node (i.e., gathered initially), (ii) an $O(\ellΔ\log(λ)+n\ell+m)$ time algorithm, if the robots start from multiple nodes (i.e., positioned arbitrarily), where $m$ is the number of edges and $Δ$ is the maximum degree of $G$, $\ell$ is the number of clusters of the robot initially and $λ$ is the maximum ID-length of the robots. Then we present a $\ln (Δ)$ approximation algorithm for the {\em minimum} dominating set which takes $O(nΔ\log (λ))$ rounds.

Aamir Hussain Chughtai, Muhammad Tahir, Momin Uppal

In this article, we consider the problem of outlier-robust state estimation where the measurement noise can be correlated. Outliers in data arise due to many reasons like sensor malfunctioning, environmental behaviors, communication glitches, etc. Moreover, noise correlation emerges in several real-world applications e.g. sensor networks, radar data, GPS-based systems, etc. We consider these effects in system modeling which is subsequently used for inference. We employ the Expectation-Maximization (EM) framework to derive both outlier-resilient filtering and smoothing methods, suitable for online and offline estimation respectively. The standard Gaussian filtering and the Gaussian Rauch-Tung-Striebel (RTS) smoothing results are leveraged to devise the estimators. In addition, Bayesian Cramer-Rao Bounds (BCRBs) for a filter and a smoother which can perfectly detect and reject outliers are presented. These serve as useful theoretical benchmarks to gauge the error performance of different estimators. Lastly, different numerical experiments, for an illustrative target tracking application, are carried out that indicate performance gains compared to similarly engineered state-of-the-art outlier-rejecting state estimators. The advantages are in terms of simpler implementation, enhanced estimation quality, and competitive computational performance.

Yu Yang, Wing Yin Cheung, Chang Liu et al.

Multiview self-supervised representation learning roots in exploring semantic consistency across data of complex intra-class variation. Such variation is not directly accessible and therefore simulated by data augmentations. However, commonly adopted augmentations are handcrafted and limited to simple geometrical and color changes, which are unable to cover the abundant intra-class variation. In this paper, we propose to extract the underlying data variation from datasets and construct a novel augmentation operator, named local manifold augmentation (LMA). LMA is achieved by training an instance-conditioned generator to fit the distribution on the local manifold of data and sampling multiview data using it. LMA shows the ability to create an infinite number of data views, preserve semantics, and simulate complicated variations in object pose, viewpoint, lighting condition, background etc. Experiments show that with LMA integrated, self-supervised learning methods such as MoCov2 and SimSiam gain consistent improvement on prevalent benchmarks including CIFAR10, CIFAR100, STL10, ImageNet100, and ImageNet. Furthermore, LMA leads to representations that obtain more significant invariance to the viewpoint, object pose, and illumination changes and stronger robustness to various real distribution shifts reflected by ImageNet-V2, ImageNet-R, ImageNet Sketch etc.

Vishal Kaushal, Ganesh Ramakrishnan, Rishabh Iyer

Submodular functions are a special class of set functions which naturally model the notion of representativeness, diversity, coverage etc. and have been shown to be computationally very efficient. A lot of past work has applied submodular optimization to find optimal subsets in various contexts. Some examples include data summarization for efficient human consumption, finding effective smaller subsets of training data to reduce the model development time (training, hyper parameter tuning), finding effective subsets of unlabeled data to reduce the labeling costs, etc. A recent work has also leveraged submodular functions to propose submodular information measures which have been found to be very useful in solving the problems of guided subset selection and guided summarization. In this work, we present Submodlib which is an open-source, easy-to-use, efficient and scalable Python library for submodular optimization with a C++ optimization engine. Submodlib finds its application in summarization, data subset selection, hyper parameter tuning, efficient training and more. Through a rich API, it offers a great deal of flexibility in the way it can be used. Source of Submodlib is available at https://github.com/decile-team/submodlib.

Subhadip Mitra

In the current connected world - Websites, Mobile Apps, IoT Devices collect a large volume of users' personally identifiable activity data. These collected data is used for varied purposes of analytics, marketing, personalization of services, etc. Data is assimilated through site cookies, tracking device IDs, embedded JavaScript, Pixels, etc. to name a few. Many of these tracking and usage of collected data happens behind the scenes and is not apparent to an average user. Consequently, many Countries and Regions have formulated legislations (e.g., GDPR, EU) - that allow users to be able to control their personal data, be informed and consent to its processing in a comprehensible and user-friendly manner. This paper proposes a protocol and a platform based on Blockchain Technology that enables the transparent processing of personal data throughout its lifecycle from capture, lineage to redaction. The solution intends to help service multiple stakeholders from individual end-users to Data Controllers and Privacy Officers. It intends to offer a holistic and unambiguous view of how and when the data points are captured, accessed, and processed. The framework also envisages how different access control policies might be created and enforced through a public blockchain including real time alerts for privacy data breach.

Bharath Sudharsan, Dineshkumar Sundaram, Pankesh Patel et al.

The majority of IoT devices like smartwatches, smart plugs, HVAC controllers, etc., are powered by hardware with a constrained specification (low memory, clock speed and processor) which is insufficient to accommodate and execute large, high-quality models. On such resource-constrained devices, manufacturers still manage to provide attractive functionalities (to boost sales) by following the traditional approach of programming IoT devices/products to collect and transmit data (image, audio, sensor readings, etc.) to their cloud-based ML analytics platforms. For decades, this online approach has been facing issues such as compromised data streams, non-real-time analytics due to latency, bandwidth constraints, costly subscriptions, recent privacy issues raised by users and the GDPR guidelines, etc. In this paper, to enable ultra-fast and accurate AI-based offline analytics on resource-constrained IoT devices, we present an end-to-end multi-component model optimization sequence and open-source its implementation. Researchers and developers can use our optimization sequence to optimize high memory, computation demanding models in multiple aspects in order to produce small size, low latency, low-power consuming models that can comfortably fit and execute on resource-constrained hardware. The experimental results show that our optimization components can produce models that are; (i) 12.06 x times compressed; (ii) 0.13% to 0.27% more accurate; (iii) Orders of magnitude faster unit inference at 0.06 ms. Our optimization sequence is generic and can be applied to any state-of-the-art models trained for anomaly detection, predictive maintenance, robotics, voice recognition, and machine vision.

Gerasimos Arvanitis

Geometry processing of 3D objects is of primary interest in many areas of computer vision and graphics, including robot navigation, 3D object recognition, classification, feature extraction, etc. The recent introduction of cheap range sensors has created a great interest in many new areas, driving the need for developing efficient algorithms for 3D object processing. Previously, in order to capture a 3D object, expensive specialized sensors were used, such as lasers or dedicated range images, but now this limitation has changed. The current approaches of 3D object processing require a significant amount of manual intervention and they are still time-consuming making them unavailable for use in real-time applications. The aim of this thesis is to present algorithms, mainly inspired by the spectral analysis, subspace tracking, etc, that can be used and facilitate many areas of low-level 3D geometry processing (i.e., reconstruction, outliers removal, denoising, compression), pattern recognition tasks (i.e., significant features extraction) and high-level applications (i.e., registration and identification of 3D objects in partially scanned and cluttered scenes), taking into consideration different types of 3D models (i.e., static and dynamic point clouds, static and dynamic 3D meshes).

Junjie Yu, Chaofeng Miao, Jun Wu et al.

Perfect optical vortices (POVs) provide an enabling solution to address the predicament induced by the strong dependence of classical optical vortices on theirs carried topological charges. Here, a type of generalized POVs with controllable impulse ring profile was proposed and demonstrated. Especially, a type of "absolute" dark POVs surrounded by two bright ringlobes in each side was presented, which provides a perfect annular potential well along those dark impulse rings for trapping steadily low-index particles, cells, or quantum gas, etc. In further, several POVs with different impulse ring profiles, including conventional POVs with bright rings, dark POVs mentioned above, and also POVs with controllable impulse ring profile, were demonstrated. This work opens up new possibilities to reshape arbitrarily the impulse ring profile for perfect vortices, and this type of novel POVs will enrich functions of optical vortices and it should be of high interest for its potential applications in optical manipulation, both quantum and classical optical communications, enhanced optical imaging, and also novel structured pumping lasers, etc..

Dhiren Kumar Basnet, Jayanta Bhattacharyya

In this article, we have defined nil clean graph of a ring $R$. The vertex set is the ring $R$, two ring elements $a$ and $b$ are adjacent if and only if $a + b$ is nil clean in $R$. Graph theoretic properties like girth, dominating set, diameter etc. of nil clean graph have been studied for finite commutative rings.

Gangli Liu, Ling Feng

Re-finding electronic documents from a personal computer is a frequent demand to users. In a simple re-finding task, people can use many methods to retrieve a document, such as navigating directly to the document's folder, searching with a desktop search engine, or checking the Recent Files List. However, when encountering a difficult re-finding task, people usually cannot remember the attributes used by conventional re-finding methods, such as file path, file name, keywords etc., the re-finding would fail. We propose a new method to support difficult re-finding tasks. When a user is reading a document, we collect all kinds of possible memory pieces of the user about the document, such as number of pages, number of images, number of math formulas, cumulative reading time, reading frequency, printing experiences etc. If the user wants to re-find a document later, we use these collected attributes to filter out the target document. To alleviate the user's cognitive burden, we use a question and answer wizard interface and provide recommendations to the answers for the user, the recommendations are generated by analyzing the collected attributes of each document and the user's experiences about them.

Andre Diatta

We investigate contact Lie groups having a left invariant Riemannian or pseudo-Riemannian metric with specific properties such as being bi-invariant, flat, negatively curved, Einstein, etc. We classify some of such contact Lie groups and derive some obstruction results to the existence of left invariant contact structures on Lie groups

Elena Kartashova

A novel discrete model (D-model) is presented describing nonlinear wave interactions in systems with small and moderate nonlinearity under narrow frequency band excitation. It integrates in a single theoretical frame two mechanisms of energy transport between modes, namely intermittency and energy cascade and gives conditions when which regime will take place. Conditions for the formation of a cascade, cascade direction, conditions for cascade termination, etc. are given and depend strongly on the choice of excitation parameters. The energy spectra of a cascade may be computed yielding discrete and continuous energy spectra. The model does not need statistical assumptions as all effects are derived from the interaction of distinct modes. In the example given -- surface water waves with dispersion function $ø^2=g\,k$ and small nonlinearity -- D-model predicts asymmetrical growth of side-bands for Benjamin-Feir instability while transition from discrete to continuous energy spectrum excitation parameters properly chosen yields the saturated Phillips' power spectrum $\sim g^2ø^{-5}$. D-model can be applied to the experimental and theoretical study of numerous wave systems appearing in hydrodynamics, nonlinear optics, electrodynamics, plasma, convection theory, etc.

Zihua Weng

The quaternion spaces can be used to describe the property of electromagnetic field and gravitational field. In the quaternion space, some coordinate transformations can be deduced from the feature of quaternions, including Lorentz transformation and Galilean transformation etc., when the coordinate system is transformed into others. And some coordinate transformations with variable speed of light can be obtained in the electromagnetic field and gravitational field.

Luis J. Boya

We review the derivation of Planck's Radiation Formula on the light of recent studies in its centenary. We discuss specially the issue of discreteness, Planck's opinion on his discovery, and the critical analysis on the contributions by Ehrenfest, Einstein, Lorentz, etc. We also address the views of T.S. Kuhn, which conflict with the conventional interpretation that discontinuty was already found by Planck.

Akito Kobayashi, Atsushi Tsuruta, Tamifusa Matsuura et al.

High-Tc superconducting materials (HTSC) have anomalous properties such as the pseudo-gap or spin-gap etc., in Hall coefficient, 1/T1T and the density of states etc. First including effects of strong on-site repulsion between d-electrons at Cu-sites, we obtain quasi-particles with super-exchange interaction Js, whose band width tends to zero, i.e., the system goes to insulator, as the hole-doping rate tends to zero. The quasi-particles correspond to Zhang-Rice singlet states. Js larger than the band width combined with the 2-dimensional character of the system induces strong antiferro-magnetic (AF) and superconducting (SC) fluctuations. We treat effects of the AF ones in the FLEX approximation and those of the SC ones in the self-consistent t-matrix approximation to show that both fluctuations in the under-doped region start to increase at T0 as T decreases from T>>Tc, the AF ones dominate the SC ones at T>Tsg, while SC ones dominate at T<Tsg. This cross-over of the fluctuations causes the anomalous phenomena in the under-doped region. We also obtain the phase diagram of HTSC consistent to one observed in experiments.

Anjan Kundu

The inhomogeneity of the media or the external forces usually destroy the integrability of a system. We propose a systematic construction of a class of quantum models, which retains their exact integrability inspite of their explicit inhomogeneity. Such models include variable mass sine-Gordon model, cylindrical NLS, spin chains with impurity, inhomogeneous Toda chain, the Ablowitz-Ladik model etc.

Luen-Chau Li

Given a classical $r$-matrix on a Poisson algebra, we show how to construct a natural family of compatible Poisson structures for the Hamiltonian formulation of Lax equations. Examples for which our formalism applies include the Benny hierachy, the dispersionless Toda lattice hierachy, the dispersionless KP and modified KP hierachies, the dispersionless Dym hierachy etc.

A. Feinstein, J. Ibáñez, R. Lazkoz

We construct exact solutions to Einstein equations which represent relativistic disks immersed into an expanding FRW Universe. It is shown that the expansion influences dynamical characteristics of the disks such as rotational curves, surface mass density, etc. The effects of the expansion is exemplified with non-static generalizations of Kuzmin-Curzon and generalized Schwarzschild disks.

R. J. Cova

New CP1-soliton behaviour on a flat torus is reported. Defined by the Weierstrass elliptic function and numerically-evolved from rest, each soliton splits up in two lumps which eventually reunite, divide and get back together again, etc.. This result invites speculation on the question of fractional topological charge.