Search

Most existing trackers based on RGB/grayscale frames may collapse due to the unreliability of conventional sensors in some challenging scenarios (e.g., motion blur and high dynamic range). Event-based cameras as bioinspired sensors encode brightness changes with high temporal resolution and high dynamic range, thereby providing considerable potential for tracking under degraded conditions. Nevertheless, events lack the fine-grained texture cues provided by RGB/grayscale frames. This complementarity encourages us to fuse visual cues from the frame and event domains for robust object tracking under various challenging conditions. In this paper, we propose a novel event feature extractor to capture spatiotemporal features with motion cues from event-based data by boosting interactions and distinguishing alterations between states at different moments. Furthermore, we develop an effective feature integrator to adaptively fuse the strengths of both domains by balancing their contributions. Our proposed module as the plug-in can be easily applied to off-the-shelf frame-based trackers. We extensively validate the effectiveness of eight trackers extended by our approach on three datasets: EED, VisEvent, and our collected frame-event-based dataset FE141. Experimental results also show that event-based data is a powerful cue for tracking. [ABSTRACT FROM AUTHOR]

With recent advancements in robotic surgery, notable strides have been made in visual question answering (VQA). Existing VQA systems typically generate textual answers to questions but fail to indicate the location of the relevant content within the image. This limitation restricts the interpretative capacity of the VQA models and their ability to explore specific image regions. To address this issue, this study proposes a grounded VQA model for robotic surgery, capable of localizing a specific region during answer prediction. Drawing inspiration from prompt learning in language models, a dual-modality prompt model was developed to enhance precise multimodal information interactions. Specifically, two complementary prompters were introduced to effectively integrate visual and textual prompts into the encoding process of the model. A visual complementary prompter merges visual prompt knowledge with visual information features to guide accurate localization. The textual complementary prompter aligns visual information with textual prompt knowledge and textual information, guiding textual information towards a more accurate inference of the answer. Additionally, a multiple iterative fusion strategy was adopted for comprehensive answer reasoning, to ensure high-quality generation of textual and grounded answers. The experimental results validate the effectiveness of the model, demonstrating its superiority over existing methods on the EndoVis-18 and EndoVis-17 datasets. [ABSTRACT FROM AUTHOR]

Exonucleases serve as efficient tools for signal processing and play an important role in biochemical reactions. Here, we identify the mechanism of cooperative exonuclease hydrolysis, offering a method to regulate the cooperative hydrolysis driven by exonucleases through the modulation of the number of bases in gap region. A signal transmission strategy capable of producing amplified orthogonal DNA signal is proposed to resolve the polarity of signals and byproducts, which provides a solution to overcome the signal attenuation. The gap-regulated mechanism combined with DNA strand displacement (DSD) reduces the unpredictable secondary structures, allowing for the coexistence of similar structures in hierarchical molecular networks. For the application of the strategy, a molecular computing model is constructed to solve the maximum weight clique problems (MWCP). This work enhances for our knowledge of these important enzymes and promises application prospects in molecular computing, signal detection, and nanomachines. A signal transmission strategy, based on the cooperative activities of Exonuclease III and Exonuclease lambda enzymes, generates discrete and tightly amplified DNA signals that overcome signal attenuation and resolve the polarity of signals and byproducts. [ABSTRACT FROM AUTHOR]

The actual needs of users for information are often hidden in multiple question answering (QA) on the same topic. In order to generate answers to users' current questions, a conversational QA system relies on getting external information from a knowledge graph (KG) and combining it with conversation history. Therefore, how to make full use of the information of KG and combining it with the conversation environment is the top priority. In a conversational knowledge graph question answering (KGQA) scenario, the follow-up questions are often incomplete and may contain a shift of focal entity. Considering and processing the constrained information which plays the key role in solving complex conversational KGQA is very important. In this paper, we propose a reinforcement learning (RL) model, which uses a dynamically maintained context entity set to capture the shift of the focal entity in the process of conversation. We then use the bidirectional encoder representations from transformers (BERT) pre-training model to obtain the semantic information from context questions and KG paths. Our model learns from not only the 1-hop path but also 2-hop path constraint of the KG at the same time and gives reward rules based on precision and certain rules, respectively. Compared with state-of-the-art methods on ConvQuestions, our model improves mean reciprocal rank (MRR) and precision at 1 (P@1) by a margin of 4.7 % and 4.3 % , respectively. Experimental results on two datasets demonstrate the effectiveness of our proposed approach. [ABSTRACT FROM AUTHOR]

Camouflaged object segmentation (COS) is a very challenging task due to the deceitful appearances of the candidate objects to the noisy backgrounds. Most existing state-of-the-art methods mimic the first-positioning-then-focus mechanism of predators, but still fail in positioning camouflaged objects in cluttered scenes or delineating their boundaries. The key reason is that their methods do not have a comprehensive understanding of the scene when they spot and focus on the objects, so that they are easily attracted by local surroundings. An ideal COS model should be able to process local and global information at the same time, i.e., to have omni perception of the scene through the whole process of camouflaged object segmentation. To this end, we propose to learn the omni perception for the first-positioning-then-focus COS scheme. Specifically, we propose an omni perception network (OPNet) with two novel modules, i.e., the pyramid positioning module (PPM) and dual focus module (DFM). They are proposed to integrate local features and global representations for accurate positioning of the camouflaged objects and focus on their boundaries, respectively. Extensive experiments demonstrate that our method, which runs at 54 fps, significantly outperforms 15 cutting-edge models on 4 challenging datasets under 4 standard metrics. The code will be made publicly available. [ABSTRACT FROM AUTHOR]

The high stability and low toxicity of bismuth-based perovskite materials can overcome current issues of lead-based perovskite solar cells (PSCs). In this study, a stable 2-amino-4-picoline (2-4py)-based bismuth perovskite material [(C6H9N2)BiI4] was developed, which has a low band gap and high stability. A one-step solution spin-coating method was used to prepare (C6H9N2)BiI4 film as the light-absorbing layer. The electron transport layers were composed of compact TiO2 (c-TiO2) and mesoporous TiO2 (mes-TiO2). Measurements by UV–vis and time-resolved photoluminescence showed that the band gap of (C6H9N2)BiI4 film was 2.0 eV and the fluorescence lifetime was 21.58 ns. The materials were assembled into a device, and comparative experimental results showed that the highest power conversion efficiency of fluorine-doped SnO2 (FTO)/c-TiO2 + mes-TiO2/(C6H9N2)BiI4/Spiro-OMeTAD/Au was 0.0328%. Although the efficiency of the current device was low, the (C6H9N2)BiI4 material exhibited excellent stability in hot and wet conditions. This indicates that (C6H9N2)BiI4 has potential for use in highly stable solar cells. [ABSTRACT FROM AUTHOR]

Theskeleton structure of human body is a natural undirected graph. Being applied to 3D body pose estimation, graph convolutional network (GCN) has achieved good results. However, the vanilla GCN ignores the differences between joints and the connections between joints with different distances. Based on the above two problems, we propose High-order Local Connection Network (HLCN) for 3D human pose estimation. On one hand, different filters for different joints are assigned to produce different weights. On the other hand, the feature of multi-hop joints synthetically is gathered into HLCN. Furthermore, we study different methods of fusing these multi-hop features and compare their performance. The new network not only takes the differences between the joints in the human skeleton into consideration, but also captures the remote dependencies between human joints. The experiment suggests that this method is superior to vanilla GCN and achieve state-of-the-art performance. The average error on the H36M dataset is 50.9 mm. [ABSTRACT FROM AUTHOR]

In recent years, deep learning has been widely applied to single image super-resolution(SISR). However, the majority of deep learning methods employ the Mean Square Error(MSE) loss as the objective optimization function, and the generated results are frequently too smooth and lack of details. In addition, the high-frequency information of the reconstructed image is severely lost, resulting in a generated image with poor visual effects. In order to address the aforementioned issues, this paper proposes a super-resolution network (RRFDB-GAN) with a receptive field module with a generative adversarial network as the main framework. The network adopts the receptive field block (RFB), which enables it to extract the features in multiple scales and to improve the discriminability. In this paper, the Residual in Residual Dense Block (RRDB) and the Residual of Receptive Field Dense Block (RRFDB) are combined into a new module, called Basic block. This module enhances the capability of feature reconstruction for low-resolution images. During the upsampling stage, a combination of Nearest Neighborhood Interpolation and Sub-pixel convolution is used to reduce the computational complexity and provide additional contextual information for super-resolution reconstruction, while achieving satisfactory performance. Finally, the four Basic blocks are integrated with the upsampling module into a simple end-to-end framework. Extensive experimental results demonstrate that the proposed method in this paper shows more details on the five test sets and outperforms other methods in terms of quantitative metrics and perception assessment. [ABSTRACT FROM AUTHOR]

In recent years, human motion prediction has become an active research topic in computer vision. However, owing to the complexity and stochastic nature of human motion, it remains a challenging problem. In previous works, human motion prediction has always been treated as a typical inter-sequence problem, and most works have aimed to capture the temporal dependence between successive frames. However, although these approaches focused on the effects of the temporal dimension, they rarely considered the correlation between different joints in space. Thus, the spatio-temporal coupling of human joints is considered, to propose a novel spatio-temporal network based on a transformer and a gragh convolutional network (GCN) (STTG-Net). The temporal transformer is used to capture the global temporal dependencies, and the spatial GCN module is used to establish local spatial correlations between the joints for each frame. To overcome the problems of error accumulation and discontinuity in the motion prediction, a revision method based on fusion strategy is also proposed, in which the current prediction frame is fused with the previous frame. The experimental results show that the proposed prediction method has less prediction error and the prediction motion is smoother than previous prediction methods. The effectiveness of the proposed method is also demonstrated comparing it with the state-of-the-art method on the Human3.6 M dataset. [ABSTRACT FROM AUTHOR]

Human motion prediction is an important research frontier, which is a key supporting technology in the fields of human–robot collaboration, automatic driving, etc. As is well known, long-term motion prediction is one most challenging direction. This paper mainly focuses on how to eliminate cumulative errors to overcome the fossilization of long-term motion sequences and aims to improve the reliability of prediction results. This paper proposed an algorithm named "fusion loss learning network," which is based on gated recurrent unit, to solve the above-mentioned problem. A fusion training method was established by combining the sampling of each step of the GRU unit with true value and output value of each previous step, which helped recover from the errors in the long-term prediction sequences. This method achieved promising results on the Human 3.6 M dataset. The results show that the proposed method could significantly improve the performance of long-term human motion prediction, and the total prediction error is reduced by 7.25% on average. [ABSTRACT FROM AUTHOR]

Computer vision (CV) technologies are assisting the health care industry in many respects, i.e., disease diagnosis. However, as a pivotal procedure before and after surgery, the inventory work of surgical instruments has not been researched with the CV-powered technologies. To reduce the risk and hazard of surgical tools' loss, we propose a study of systematic surgical instrument classification and introduce a novel attention-based deep neural network called SKA-ResNet which is mainly composed of: (a) A feature extractor with selective kernel attention module to automatically adjust the receptive fields of neurons and enhance the learnt expression and (b) A multi-scale regularizer with KL-divergence as the constraint to exploit the relationships between feature maps. Our method is easily trained end-to-end in only one stage with few additional calculation burdens. Moreover, to facilitate our study, we create a new surgical instrument dataset called SID19 (with 19 kinds of surgical tools consisting of 3800 images) for the first time. Experimental results show the superiority of SKA-ResNet for the classification of surgical tools on SID19 when compared with state-of-the-art models. The classification accuracy of our method reaches up to 97.703%, which is well supportive for the inventory and recognition study of surgical tools. Also, our method can achieve state-of-the-art performance on four challenging fine-grained visual classification datasets. [ABSTRACT FROM AUTHOR]

Three-dimensional human motion synthesis is one of the key technologies in the field of computer animation and multimedia applications. It is well known that the human body's own motion is full of strong personality, emotion, and high-dimensional characteristics, leading to the automatic synthesis of diverse and lifelike 3D human motion data continues to be a challenging task. Facing the challenge, this paper proposes a human motion synthesis framework based on hierarchical learning recurrent neural networks (HL-RNN). The framework includes a low-level network and a high-level network, which are used to extract the path information of the movement and the spatio-temporal relationship of the human bone structure, respectively. Then, after fusion, motions that satisfy the path constraints could be generated. This method can not only synthesize high-quality human movements that follow a specified trajectory, but also synthesize smooth transitions between various movements, and can also be used to synthesize data of different motion styles. Compared with some latest methods, experiments showed that the proposed method can significantly improve the quality and generalization performance of motion synthesis. [ABSTRACT FROM AUTHOR]

Inappropriate packages and packaging materials are the main causes of bruise damage in the postharvest handling of kiwifruit. This study was conducted to identify the suitable materials to be used as collection box during sorting and grading of kiwifruits. Three different materials of box, wooden box, high-density polyethylene (HDPE) box, and expanded polystyrene (EPS) box were used to simulate kiwifruit dropping in grading line. Among three treatments, the lowest weight loss (4.6%) was observed in the EPS box surface-impacted fruits, the bruise index and bruise area of fruits from EPS box were 36.2% and 47.1% less than that of wooden box, respectively. Furthermore, the peak level of ethylene production and respiration rate of fruits from EPS box was 32.6%, 28.9% lower than that of wooden box, respectively. Compared to wooden box and HDPE box surface, EPS box surface slowed softening, reduced electrolyte leakage and malondialdehyde accumulation. Additionally, lower bruise damage by the EPS box surface was also displayed in physiological attributors including hydrogen peroxide (H2O2), superoxide anion (O2–) and the activities of antioxidant enzymes. The collective data indicated that the EPS box surface reduced the adverse physiological changes caused by dropping of kiwifruit. These results indicated the potential of EPS as packaging material of collection box in grading line to reduce bruise damage and to preserve the quality of kiwifruit. [ABSTRACT FROM AUTHOR]

This article investigates the passivity of reaction-diffusion genetic regulatory networks (GRNs) with time-varying delays and uncertainty terms under Dirichlet, Neumann, and Robin boundary conditions. We provide delay-dependent stability criteria by constructing appropriate Lyapunov-Krasovskii functions and linear matrix inequalities, and offer conditions sufficient to ensure the passivity of GRNs. We conducted a comparative analysis of GRNs under these three conditions. Numerical examples of the proposed approaches are provided to illustrate its effectiveness, and represent the three-dimensional figures of the trajectories of the concentrations of mRNA and the proteins of GRNs under Dirichlet boundary conditions. [ABSTRACT FROM AUTHOR]

The super-resolution (SR) problem is challenging due to the diversity of image types with little shared properties as well as the speed required by online applications, e.g., target identification. In this paper, we explore the merits and demerits of recent deep learning-based and conventional patch-based SR methods and show that they can be integrated in a complementary manner, while balancing the reconstruction quality and time cost. Motivated by this, we further propose an integration framework to take the results from FSRCNN and A+ methods as inputs and directly learn a pixel-wise mapping between the inputs and the reconstructed results using the Gaussian conditional random fields. The learned pixel-wise integration mapping is flexible to accommodate different upscaling factors. Experimental results show that the proposed framework can achieve superior SR performance compared with the state of the arts while being efficient. [ABSTRACT FROM AUTHOR]

We have analyzed the stability of interactional genetic regulatory networks with reaction-diffusion terms under Dirichlet boundary conditions in this article. Corresponding to interaction between unstable genetic regulatory networks and stable genetic regulatory networks, the model is given, and a stability criterion is proposed through construction of appropriate Lyapunov-Krasovskii functions and linear matrix inequalities (LMI). By means of a numerical simulation, we have proved the effectiveness and correctness of the theorem, and we analyzed the factors that influence the stability for interactional genetic regulatory networks. [ABSTRACT FROM AUTHOR]

In this paper, we present a parameter learning method to reflect the rapidly changing behaviors in the traffic flow simulation process, in which we insert virtual vehicles into the real trajectory data. We come up with a real-virtual interaction model and then we use genetic algorithm to learn some parameters in the model with the purpose to get some specific driving characteristics. Then we propose a real-virtual interaction system to vividly simulate the various interaction behaviors between the real vehicles and the virtual ones. Our results are compared to the existing methods to prove the effectiveness of our presented method. [ABSTRACT FROM AUTHOR]

In this study, we propose a practical approach for calculating the analytical inverse kinematic solution for a seven-degrees of freedom (7-DOF) space manipulator with joint and attitude limits. Instead of utilizing traditional velocity-based approaches that limit the ranges of joints by calculating the velocity-level Jacobian matrix, we propose a position-based approach for evaluating the ranges of feasible inverse kinematic solutions. We then search for the optimal solution, which is estimated based on the disturbance that acts on the base of the manipulator to obtain the final solution. First, the concept of the redundancy of manipulators is defined and each joint is parameterized by the redundancy. Second, how the joint limits affect this redundancy is discussed. Third, a practical approach (include the objective function that the author needs to minimize) is proposed for dealing with the inverse kinematic problem of 7-DOF manipulators. Finally, the validity of this approach is verified by numerical simulation. [ABSTRACT FROM AUTHOR]

A new improved hybrid optimization algorithm - PGATS algorithm, which is based on toy off-lattice model, is presented for dealing with three-dimensional protein structure prediction problems. The algorithm combines the particle swarm optimization (PSO), genetic algorithm (GA), and tabu search (TS) algorithms. Otherwise, we also take some different improved strategies. The factor of stochastic disturbance is joined in the particle swarm optimization to improve the search ability; the operations of crossover and mutation that are in the genetic algorithm are changed to a kind of random liner method; at last tabu search algorithm is improved by appending a mutation operator. Through the combination of a variety of strategies and algorithms, the protein structure prediction (PSP) in a 3D off-lattice model is achieved. The PSP problem is an NP-hard problem, but the problem can be attributed to a global optimization problem of multi-extremum and multi-parameters. This is the theoretical principle of the hybrid optimization algorithm that is proposed in this paper. The algorithm combines local search and global search, which overcomes the shortcoming of a single algorithm, giving full play to the advantage of each algorithm. In the current universal standard sequences, Fibonacci sequences and real protein sequences are certified. Experiments show that the proposed new method outperforms single algorithms on the accuracy of calculating the protein sequence energy value, which is proved to be an effective way to predict the structure of proteins. [ABSTRACT FROM AUTHOR]

For the existing motion capture (MoCap) data processing methods, manual interventions are always inevitable, most of which are derived from the data tracking process. This paper addresses the problem of tracking non-rigid 3D facial motions from sequences of raw MoCap data in the presence of noise, outliers and long time missing. We present a novel dynamic spatiotemporal framework to automatically solve the problem. First, based on a 3D facial topological structure, a sophisticated non-rigid motion interpreter (SNRMI) is put forward; together with a dynamic searching scheme, it cannot only track the non-missing data to the maximum extent but recover missing data (it can accurately recover more than five adjacent markers under long time (about 5 seconds) missing) accurately. To rule out wrong tracks of the markers labeled in open structures (such as mouth, eyes), a semantic-based heuristic checking method was raised. Second, since the existing methods have not taken the noise propagation problem into account, a forward processing framework is presented to solve the problem. Another contribution is the proposed method could track facial non-rigid motions automatically and forward, and is believed to greatly reduce even eliminate the requirements of human interventions during the facial MoCap data processing. Experimental results proved the effectiveness, robustness and accuracy of our system. [ABSTRACT FROM AUTHOR]

MoCap (motion capture)-based animation is a hot issue in computer animation research currently. Based on the optical MoCap system, this paper proposes a novel cross-mapping based facial expression simulating method. To overcome the problem of the false upper and lower jaw correlation derived from the facial global RBF-based cross-mapping method, we construct a functional partition based RBF cross-mapping method. During model animating, enhanced markers are added and animated by our proposed skin motion mechanism. In addition, based on the enhanced markers, an improved RBF-based animating approach is raised to derive realistic facial animation. Further more, a pre-computing algorithm is presented to reduce computational cost for real-time simulation. The experiments proved that the method can not only map the MoCap data of one subject to different personalized faces but generate realistic facial animation. [ABSTRACT FROM AUTHOR]

The problem of protein structure prediction in the hydrophobic-polar (HP) lattice model is the prediction of protein tertiary structure. This problem is usually referred to as the protein folding problem. This paper presents a method for the application of an enhanced hybrid search algorithm to the problem of protein folding prediction, using the three dimensional (3D) HP lattice model. The enhanced hybrid search algorithm is a combination of the particle swarm optimizer (PSO) and tabu search (TS) algorithms. Since the PSO algorithm entraps local minimum in later evolution extremely easily, we combined PSO with the TS algorithm, which has properties of global optimization. Since the technologies of crossover and mutation are applied many times to PSO and TS algorithms, so enhanced hybrid search algorithm is called the MCMPSO-TS (multiple crossover and mutation PSO-TS) algorithm. Experimental results show that the MCMPSO-TS algorithm can find the best solutions so far for the listed benchmarks, which will help comparison with any future paper approach. Moreover, real protein sequences and Fibonacci sequences are verified in the 3D HP lattice model for the first time. Compared with the previous evolutionary algorithms, the new hybrid search algorithm is novel, and can be used effectively to predict 3D protein folding structure. With continuous development and changes in amino acids sequences, the new algorithm will also make a contribution to the study of new protein sequences. [ABSTRACT FROM AUTHOR]

A rounding error analysis for the symplectic Lanczos method is given to solve the large-scale sparse Hamiltonian eigenvalue problem. If no breakdown occurs in the method, then it can be shown that the Hamiltonian structure preserving requirement does not destroy the essential feature of the nonsymmetric Lanczos algorithm. The relationship between the loss of J-orthogonality among the symplectic Lanczos vectors and the convergence of the Ritz values in the symmetric Lanczos algorithm is discussed. It is demonstrated that under certain assumptions the computed J-orthogonal Lanczos vectors lose the J-orthogonality when some Ritz values begin to converge. Our analysis closely follows the recent works of Bai and Fabbender. [ABSTRACT FROM AUTHOR]

Sticker model is one of the basic models in the DNA computer models. This model is coded with single-double stranded DNA molecules. It has the following advantages that the operations require no strands extension and use no enzymes; What's more, the materials are reusable. Therefore, it arouses attention and interest of scientists in many fields. In this paper, we extend and improve the sticker model, which will be definitely beneficial to the construction of DNA computer. This paper is the second part of our series paper, which mainly focuses on the application of sticker model, it mainly consists of the following three sections: the matrix representation of sticker model is first presented; then a brief review of the past research on graph and combinatorial optimization, such as the minimal set covering problem, the vertex covering problem, Hamiltonian path or cycle problem, the maximal clique problem, the maximal independent problem and the Steiner spanning tree problem, is described; Finally a DNA algorithm for the graph isomorphic problem based on the sticker model is given. [ABSTRACT FROM AUTHOR]

Sticker model is one of the basic models in the DNA computer models. This model is coded with single-double stranded DNA molecules. It has the following advantages that the operations require no strands extension and use no enzymes; What's more, the materials are reusable. Therefore it arouses attention and interest of scientists in many fields. In this paper, we will systematically analyze the theories and applications of the model, summarize other scientists' contributions in this field, and propose our research results. This paper is the theoretical portion of the sticker model on DNA computer, which includes the introduction of the basic model of sticker computing. Firstly, we systematically introduce the basic theories of classic models about sticker computing; Secondly, we discuss the sticker system which is an abstract computing model based on the sticker model and formal languages; Finally, extend and perfect the model, and present two types of models that are more extensive in the applications and more perfect in the theory than the past models: one is the so-called k-bit sticker model, the other is full-message sticker DNA computing model. [ABSTRACT FROM AUTHOR]

In this study, we propose two techniques for clustering genetic regulatory networks with mixed delays. Laws for identifying coupling parameters are designed, and the Lyapunov theorem and Lipschitz condition are employed to achieve the desired stability and to identify the coupling parameters. Using our proposed methods, the value of a cluster or the sum of clusters can approach the expected values. Numerical examples verify the feasibility and effectiveness of this method. [ABSTRACT FROM AUTHOR]