Recent research endeavors have shown the potential of using feed-forward convolutional neural networks to accomplish fast style transfer for images. In this work, we take one step further explore possibility exploiting a network perform videos and simultaneously maintain temporal consistency among stylized video frames. Our is trained by enforcing outputs consecutive frames be both well temporally consistent. More specifically, hybrid loss proposed capitalize on content information input...

Abstract Significant progress is made in the treatment of metastatic colorectal cancer (mCRC) patients, however, therapeutic options remain limited for patients with mCRC. In recent years, traditional Chinese medicine (TCM) has gained significant attention. Among these, Huangqin Houpo decoction demonstrated efficacy mCRC treatment. Despite its promise, active ingredients and mechanisms underlying anticancer effects unclear. Using integrative pharmacological approaches, six compounds are...

Many systems, such as distributed operating complex networks, and high throughput web-based applications, are continuously generating large volume of event logs. These logs contain useful information to help system administrators understand the running status pinpoint failures. Generally, due scale complexity modern generated beyond analytic power human beings. Therefore, it is imperative develop a comprehensive log analysis support effective management. Although number mining techniques...

The Location-Based Social Networks (LBSN) (e.g., Facebook, etc.) have many attributes ratings, reviews, that play a crucial role for the Point-of-Interest (POI) recommendations. Unlike reviews can help users to elaborate their consumption experience in terms of relevant factors interest (aspects). Though some existing systems exploited user most them are less transparent and non-interpretable (as they conceal reason behind recommendation). These reasons motivated us towards explainable...

Gastric cancer has become a global health issue, severely disrupting daily life. Early detection in gastric patients and immediate treatment contribute significantly to the protection of human health. However, routine examinations carry risk complications are time-consuming. We proposed framework predict non-invasively conveniently. A total 703 tongue images were acquired using bespoke image capture instrument, then dataset containing subjects with without was created. As by this instrument...

The blind source separation (BSS) problem with an unknown number of sources is important practical issue that usually skipped by assuming the n known and equal to m sensors. This letter studies general BSS satisfying ≥ n. First, it shown mutual information outputs network a cost function for BSS, provided mixing matrix full column rank m×m separating nonsingular. reaches its local minima at points, where consist desired signals m−n redundant signals. Second, proved natural gradient algorithm...

This paper studies the joint estimation problem of ranges, DOAs, and frequencies near-field narrowband sources proposes a new computationally efficient algorithm, which employs symmetric uniform linear array, uses eigenvalues together with corresponding eigenvectors two properly designed matrices to estimate signal parameters, does not require searching for spectral peak or pairing among parameters. In addition, proposed algorithm can be applied in arbitrary Gaussian noise environment since...

Blind source separation (BSS) aims at recovering statistically independent signals from their linear mixtures without knowing the mixing coefficients. Besides component analysis, nonlinear principal analysis (NPCA) is shown to be another useful tool for solving this problem, but it requires that measured data prewhitened. By taking into account autocorrelation matrix of data, we present in paper a modified NPCA criterion, and develop least-mean-square (LMS) algorithm recursive least-squares...

For the sake of enhancing identification ability current network and meeting needs high accuracy distinguishing similar small objects (foliage) in complex scenes, this paper proposes a modified region-based fully convolutional which adopts Inception V3 accompanying with residual connection as main framework. Incorporating deep learning module into can not only save computational cost by factorising convolutions, but also mitigate vanishing gradients causing increasing depth network....

Abstract Accurately identifying tropical fish serves as a crucial indicator, offering an insight into the state of marine biodiversity and condition coral reef ecosystems. However, current detection networks are prone to omission misidentification due occlusion between complex underwater environment. This paper proposes modified approach named FishFocusNet, in which alterable kernel convolution modules, asymptotic feature pyramid network (AFPN), Shape‐IoU integrated YOLOv8. To extract more...

The Point-of-Interest (POI) preference of a user varies by locality, item type, and the co-visitors, e.g., user1 user2 can have closest on food items but not historic sites, etc. A locality different trends (e.g., popular for food, recreation, etc.) user's span across multiple such trends. good recommender should also exploit aggregated Most existing studies group category or global preferences which might be relevant locality-based preferences. We propose HiCaPS (<u>Hi</u>erarchical...

The Intercell Interference (ICI) problem is one of the main challenges in Long Term Evolution (LTE) downlink system.In order to deal with ICI problem, this paper proposes a joint resource block and transmit power allocation scheme LTE networks.The proposed implemented three phases: (1) priority users calculated based on interference level, Quality Service (QoS) Head Line (HoL) delay;(2) each cell are scheduled specified subbands their priority; (3) eNodeBs dynamically control using fuzzy...

The revolution of World Wide Web (WWW) and smart-phone technologies have been the key-factor behind remarkable success social networks. With ease availability check-in data, location-based networks (LBSN) (e.g., Facebook1, etc.) heavily explored in past decade for Point-of-Interest (POI) recommendation. Though many POI recommenders defined, most them focused on recommending a single location or an arbitrary list that is not contextually coherent. It has cumbersome to rely such systems when...

For joint blind channel estimation and multiuser detection in direct-sequence code division multiple access (DS-CDMA) systems, a combined scheme is proposed, which couples the recursive least-squares (RLS) adaptations PASTd subspace tracking algorithm. A major advantage of RLS-PASTd algorithm that it admits an adaptive implementation with low computational complexity instead singular value decomposition. Simulation results demonstrate fast capability high steady-state accuracy proposed both...

In statistical signal processing, the sequential Monte Carlo (SMC) method is powerful and can approach theoretical optima. However, its computational complexity usually very high, especially in multiple-input multiple-output (MIMO) systems. This paper presents a new low-complexity SMC (LC-SMC) algorithm for blind detection MIMO systems, main idea of which to shrink sampling space via channel estimation initialized using first differentially modulated symbol then updated samples. Since...

In this letter, the problem of blind separation n independent sources from their m linear instantaneous mixtures is considered. First, a generalized contrast function defined as valuable extension existing classical and nonsymmetrical functions. It applicable to overdetermined (m > n) with an unknown number sources, because not only components but also redundant ones are allowed in outputs system. Second, natural gradient learning algorithm developed primarily for complete case = shown work...

Face alignment involves locating several facial parts such as eyes, nose and mouth, has been popularly tackled by fitting deformable models. In this paper, we explore the effect of combination structured random regressors Constrained Local Models (CLMs). Unlike most previous CLMs, proposed a novel to give joint prediction rather than pursuing independence while learning response map for each part. our method, first present fast algorithm learn local graph, which will then be efficiently...