Traffic sign detection has made great progress with the rise of deep learning in recent years. As a result of the complex and changeable traffic environment, detecting small traffic signs in a real-world scene is still a challenging problem. In this paper, a novel group multi-scale attention pyramid network is proposed to address the problem. Specifically, to aggregate the feature at different scales

Hindsight Experience Replay (HER) is a multi-goal reinforcement learning algorithm for sparse reward functions. The algorithm treats every failure as a success for an alternative (virtual) goal that has been achieved in the episode. Virtual goals are randomly selected, irrespective of which are most instructive for the agent. In this paper, we present two improvements over the existing HER algorithm

Learning to infer missing links is one of the fundamental tasks in the knowledge graph. Instead of reasoning based on separate paths in the existing methods, in this paper, we propose a new model, Sequential Relational Graph Convolutional Network (SRGCN), which treats the multiple paths between an entity pair as a sequence of subgraphs. Specifically, to reason the relationship between two entities

Knowledge graph (KG) embedding projects the graph into a low-dimensional space and preserves the graph information. An essential part of a KG is the ontology, which always is organized as a taxonomy tree, depicting the type (or multiple types) of each entity and the hierarchical relationships among these types. The importance of considering the ontology during KG embedding lies in its ability to provide

Emotion recognition in conversation (ERC) is an important research topic in artificial intelligence. Different from the emotion estimation in individual utterances, ERC requires proper handling of human interactions in conversations. Several approaches have been proposed for ERC and achieved promising results. In this paper, we propose a correction model for previous approaches, called “Dialogical

This paper investigates the reconstructibility of switched Boolean control networks (SBCNs). Several new types of definition about reconstructibility are proposed where the existence and arbitrariness of the switching signal and the input sequence are inconsistent. A weighted pair graph describing all pairs of states which are indistinguished is defined. A deterministic finite automata method is applied

Face expression recognition (FER) on low expression intensity is not well studied in the literature. This paper investigates this problem and presents a novel Generative Adversarial Network (GAN) based multimodal approach to it. The method models the tasks of intensity enhancement and expression recognition jointly, ensuring that the synthesize faces not only present expression of high intensity, but

Salient object detection simulates the attention mechanism of human behavior to grasp the most attractive objects in the images. Recently edge information has been introduced to enhance the sharp contour in RGB image saliency detection. Inspired by it, we probe into the edge-guided RGB-D image saliency detection. There are two key problems need to be solved. One is how to extract edge information from

Artificial Neural Networks (ANNs) achieve high accuracy in various cognitive tasks (i.e., inferences), but often fail to meet power and latency budgets due to intensive computational overheads. To address the challenge, Spiking Neural Networks (SNNs) have emerged as high-performance and power-efficient alternatives thanks to their theoretically efficient spike-driven computations. The spike-based computations

Weakly Supervised Object Localization (WSOL) is a technique for obtaining the object location from attention maps of the classification network, without using bounding box annotations. Existing WSOL approaches lack the modeling of the correlations between different regions of the target object. Hence they can only locate some discriminative attentions, which are small and sparse. Besides, they introduce

Domain alignment (DA) has been widely used in unsupervised domain adaptation. Many existing DA methods assume that a low source risk, together with the alignment of distributions of source and target, means a low target risk. In this paper, we show that this does not always hold. We thus propose a novel metric-learning-assisted domain adaptation (MLA-DA) method, which employs a novel triplet loss for

This paper formulates new theoretical results concerning the multiple O(t-q) stability and instability for a class of time-varying delayed fractional-order Cohen-Grossberg neural networks (FoCGNNs) with Gaussian activation functions. With the aid of geometrical configurations obtained from the FoCGNNs model and Gaussian functions, the state space are partitioned into 3k subspaces, where k is a nonnegative

Crowd counting, i.e., estimating the number of people in crowded areas, has attracted much interest in the research community. Although many attempts have been reported, crowd counting remains an open real-world problem due to the vast density variations and severe occlusion within the interested crowd area. In this paper, we propose a novel Pyramid Density-Aware Attention based network, abbreviated

This paper proposes a hybrid architecture based on neural networks, fuzzy systems, and n-uninorms for solving pattern classification problems, termed as ENFS-Uni0 (short for evolving neuro-fuzzy system based on uni-nullneurons). The model can produce knowledge in an on-line (single-pass) and evolving learning context in a particular form of neuro-fuzzy rules representing the dependencies among input

Many methods achieve the visual object tracking task with deep learning technologies. As the deep features of different levels contain various semantic information and functions, this paper presents a multi-hierarchy feature aggregation approach to tackle the specific issues in the tracking task, which consists of two aspects. On one hand, this paper integrates the features captured by the offline

Recently, deep convolutional neural networks have demonstrated great effectiveness in action recognition with both RGB and optical flow in the past decade. However, existing studies generally treat all frames and pixels equally, potentially leading to poor robustness of models. In this paper, we propose a novel parameter-free spatial–temporal pooling block (referred to as STP) for action recognition

Single-molecule localization microscopy is a widely used technique in biological research for measuring the nanostructures of samples smaller than the diffraction limit. This study uses multifocal plane microscopy and addresses the three-dimensional (3D) single-molecule localization problem, where lateral and axial locations of molecules are estimated. However, when multifocal plane microscopy is used

Accurate traffic speed forecasting is critical in advanced transportation management and traveler route planing. Considering the important influences of spatial–temporal factors and excellent performance of recurrent neural networks (RNNs) in the field of time series analyzing, in this paper, the features injected recurrent neural networks (FI-RNNs) were proposed, which combines sequential time data

When using a Bayesian network to model a practical problem, weak prior knowledge projected as ancestral constraints is necessary. However, it is difficult to directly utilize these non-decomposable constraints using search strategies based on the decomposable score. In this study, we attempt to solve this problem by conducting an implicate path-space search graph and driving the A* algorithm, which

In the process of Human-robot skill transfer, we require the robot to reproduce the trajectory of teacher and expect that the robot can generalize the learned trajectory. For the trajectory after generalization, we expect that the robot arm can accurately track. However, because the model of the robot can not be accurately obtained, some researchers have proposed using a neural network to approximate

We study a task of aligning time-sync video comments (danmaku) to narrative video storylines, which is helpful for finding semantic segmentation of videos and conducting fine-grained user feedback analyses. Due to the mismatch of text styles and the shift of topics, it is hard to apply previous semantic matching models directly for the alignment. To tackle the problem, we propose to utilize variational

Unstructured deep neural network (DNN) pruning have been widely studied. However, previous schemes only focused upon compressing the model’s memory footprint, which had led to relatively low reduction ratio in computational workload. This study demonstrates that the main reason behind is the inconsistent distribution of memory footprint and workload of the DNN model among different layers. Based on

With the supplement of texture and geometry cues in depth maps, some difficult scenes of salient object detection (SOD) in 2D images can be overcome. However, some distractors in the depth maps with relatively poor quality may interfere with SOD. Thus, how to suppress the interference of depth maps and extract valuable depth cues, is a critical issue to serve as effective complements to RGB cues. Aiming

This paper considers the fixed-time tracking control problem for system-constrained nonlinear pure-feedback systems involving input saturation and output constraints. A sequence of auxiliary virtual and actual input signals is designed to obtain an expression for the system tracking error and stabilize the system. A combination of the fixed-time stability theory, barrier Lyapunov function, and radial

Representation learning (RL) has gained increasingly considerable attention in intelligent fault diagnosis due to its great capability of automatically learning useful features. Most existing studies focus on developing various variants about RL through modifying loss function of original versions, whereas it is a challenging task. Using the promising sparse filtering as the basic module, this paper

Recently, convolutional neural networks have improved the results of salient object detection by a significant margin. Most existing methods focus on aggregating multi-level features but pay little attention to the differences between their spatial resolution. Besides, the widely used binary cross entropy loss treats all pixels equally and neglects the fact that pixels near the salient boundaries are

Generative adversarial networks (GANs) have recently achieved impressive results in facial age synthesis. However, these methods usually select an autoencoder-style generator. And the bottleneck layer in the encoder-decoder generally gives rise to blurry and low-quality generation. To address this limitation, we propose a novel attention-aware conditional generative adversarial network (ACGAN). First

DeepLab—one of the most critical deep neural network models for image segmentation—has achieved the most advanced performance in the field of image semantic understanding. However, the rich parameters and calculation of deep convolutional neural networks (DCNNs) demand further research on neuro-inspired computing chips specifically designed for hardware acceleration and AI applications. This motivates

Most of the excellent methods for visual object tracking are based on RGB videos. With the popularity of depth cameras, the research on RGB-D(RGB+depth) tracking has gradually gained extensive attention. The depth map provides more available information for dealing with intractable tracking problems. How to make full use of depth maps to construct a better tracker is the foremost problem to be settled

Identifying adverse drug reaction (ADR) entities from texts is a crucial task for pharmacology, and it is the basis for the ADR relation extraction task. The publicly available resources on this task include PubMed abstracts, social media, and other resources. Among these resources, social media data can reflect the reactions of drug users after taking medicine in real-time and update quickly. However

When the training data and the test data do not obey the same distribution, the performances of many object detection methods always decrease greatly. Naturally, domain adaptation methods at feature level are proposed. The basic idea is to adapt the feature extraction network such that the feature distributions of the source and target domains match. We propose a new method that is built directly on

This paper proposed a novel approach to generate face caricatures automatically from a single portrait image. We decompose the process of 3D face caricatures generation into two independent subtasks: appearance transfer of texture and the geometry transfer of mesh. For the appearance transfer, we design a GAN-based network named CariFaceGAN to learn the style mapping from portrait to caricature, in

Adaptive formation tracking problems for swarm systems with multiple leaders and switching topologies are studied in this paper. It is required that the followers form time-varying formations and track the positions of the leaders simultaneously, where the bio-inspired adaptive formation tracking control method is originated from the grey wolves with strict hierarchy and flexible communication structure

As the size of dataset increases, it has become a difficult task to explore structures of interest from crowded parallel coordinates due to visual clutter. Numerous methods have been proposed to simplify the visualization of crowded parallel coordinates, such as filtering, bundling and sampling. However, contextual structures are hardly preserved in the course of simplification, which make significant

In self-attention networks (SANs), positional embeddings are used to model order dependencies between words in the input sentence and are added with word embeddings to gain an input representation, which enables the SAN-based neural model to perform (multi-head) and to stack (multi-layer) self-attentive functions in parallel to learn the representation of the input sentence. However, this input representation

Web search has become a popular way for people to obtain information. Due to the complexity of search task, one retrieval cannot serve all of user’s information needs. Multiple interactions with the search engine are required, and session search comes into being. Currently proposed session search methods still cannot take advantage of interactive information to capture the user’s implicit information

Compared with the classic object detection problem, detecting objects in aerial images has some special challenges including huge orientation variations, complicated and large background, and wide multi-scale distribution. Considering these three challenges together, we propose a novel arbitrary-oriented object detection framework consisting of three main parts. Firstly, the Cascading Attention Network

Urban hotspots spatiotemporal prediction is a long-term but challenging task for urban management and smart city construction. Accurate urban hotspots spatiotemporal prediction can improve urban planning, scheduling and, security capability, reduce resource consumption. Existing deep spatiotemporal prediction methods mainly utilize geographic grid based image, some given network structure or some additional

Deep convolutional neural networks have gained aggressive success in salient object detection. This paper uses the Multi-Scale Feature Extraction Module (MFEM) for each backbone level to get multi-scale contextual knowledge. We propose the Cross Feature Aggregation Modules(CFAM) to integrate the various features from adjacent levels, which comparatively propagate less noise due to small up-/down sampling

Deep-layer autoencoder (DAE) provides a powerful way for medical image analysis, while it remains a daunting challenge due to the limited samples but high dimension. In this paper, a DAE with sparse and graph Laplacian regularization, termed as GSDAE, is presented to identify significant differences of dynamic functional connectivity (dFC) between child and young adult groups. The proposed model incorporates

The skeletal data has been an alternative for the human action recognition task as it provides more compact and distinct information compared to the traditional RGB input. However, unlike the RGB input, the skeleton data lies in a non-Euclidean space that traditional deep learning methods are not able to use their fullest potential. Fortunately, with the emerging trend of Geometric deep learning, the

The recent achievements claimed that the fly visual neural system could serve a type of artificial computation model for revealing the properties of fly’s learning, memory and decision-making. It, however, still keeps open how to borrow the properties to develop artificial visual neural network optimization models. Hereby, inspired by gradient descent, fly’s visual information-processing and innate

The relationships between muscle movements and neural signals make it possible to decode silent speech based on neuromuscular activities. The decoding can be formulated as a supervised classification task. The electromyography (EMG) captured from surface articulatory muscles contains useful information that can help assist in decoding of speech. Spectrograms obtained from EMG have a wealth of information

Deep convolutional neural networks (CNNs) is difficult to deploy on resource-constrained devices due to its huge amount of computation. Channel pruning is an effective method to reduce the amount of computation and accelerate network inference. Most of channels pruning methods use statistics from a single structure (convolutional layer or batch normalization layer) of the sparse network to evaluate

In this study, we investigated a means to improve the robustness of deep network training on visual recognition tasks without sacrificing accuracy. The contribution of this work can reduce the dependence on model decay to gain a strong defense against malicious attacks, especially from adversarial samples. There are two major challenges in this study. First, the model defense capability should be strong

Anchor mechanism is widely applied in scene text detection methods and demonstrates promising performance. However, existing anchor mechanisms have two major limitations, namely handcrafted anchor design and hard-wired anchor assignment. We propose a novel Grouped Channels Composition(GCC) block to achieve the data-driven anchor design and adaptive anchor assignment. To be more specific, our GCC block

In this paper, we propose an attention-guided multi-scale fusion network (named as AMS-Net) for crowd counting in dense scenarios. The overall model is mainly comprised by the density and the attention networks. The density network is able to provide a coarse prediction of the crowd distribution (density map), while the attention network helps to distinguish crowded regions from backgrounds. The output

Object detection based on deep learning typically relies on a large number of training data, which may be very labor-consuming to prepare. In this paper, we attempt to tackle the problem by addressing the One-Shot Object Detection (OSOD) task. Given a novel image denoted as the query image whose category label is not included in the training data, OSOD aims to detect objects of the same class in a

Line matching plays an essential role in Structure from Motion (SFM) and Simultaneous Localization and Mapping (SLAM), especially in low-texture scenes, where feature points are hard to be detected. In this paper, we present a new method by combining Convolutional Neural Networks and Graph Convolutional Networks to match line segments in pairs of images. We design a graph-based method to predict the

Unsupervised domain adaptive (UDA) person re-identification (re-ID) aims to adapt the model trained on a labeled source domain to an unlabeled target domain. For pseudo-label-based UDA methods, pseudo labels noise is the main problem for model degradation and the factors that cause noise are complex. In this paper, a novel learning with noisy labels (LNL) method for UDA person re-ID is proposed to

Semantic segmentation has been a major topic in research and industry in recent years. However, due to the computation complexity of pixel-wise prediction and backpropagation algorithm, semantic segmentation has been demanding in computation resources, resulting in slow training and inference speed and large storage space to store models. Existing schemes that speed up segmentation network change the

Data-driven fault diagnostics of safety-critical systems often faces the challenge of a complete lack of labeled data from faulty system conditions at training time. Since faults of unknown types can arise during deployment, fault diagnostics in this scenario is an open-set learning problem. Without labels and samples from the possible fault types, the open-set diagnostics problem is typically reformulated

This paper introduces the idea of learning uniformly ergodic Markov chain for one-against-all support vector machine (OAA-SVM) algorithm. We first obtain the generalization error of OAA-SVM with fast learning rate for uniformly ergodic Markov samples. We also propose a new OAA-SVM method with Markov sampling (OAA-SVM-MS). The experimental researches for benchmark repository confirm that the OAA-SVM-MS

Although recent deep generative models are able to generate high-resolution, diverse natural samples from complex datasets, the generated samples still exist some problems in terms of images structure and detailed texture. In this paper, we propose a novel network architecture–SEDA-GAN that can learn the potential relationship in the dimension of the channel to enhance the generation performance of

Tensor decomposition is an effective approach to compress over-parameterized neural networks and to enable their deployment on resource-constrained hardware platforms. However, directly applying tensor compression in the training process is a challenging task due to the difficulty of choosing a proper tensor rank. In order to address this challenge, this paper proposes a low-rank Bayesian tensorized

Gait recognition is a biometric technology used to identify people from their walking patterns. Most traditional Computer Vision-based gait recognition approaches have been tested to work satisfactorily given similar training and test conditions. However, there may be situations where the two conditions differ due to the presence of co-variate objects like bag, coat, and/or others. The presence of

It is expensive and time-consuming to collect sufficient labeled data for human activity recognition (HAR). Domain adaptation is a promising approach for cross-domain activity recognition. Existing methods mainly focus on adapting cross-domain representations via domain-level, class-level, or sample-level distribution matching. However, they might fail to capture the fine-grained locality information

Document image classification plays a vital role in the document image processing system. Thus it is of great importance to have a clear understanding of the state-of-the-art of the document image classification field, especially in this deep learning era, which will facilitate the development of effective document image processing systems. In this paper, we provide a comprehensive survey of the progress

Due to boundary ambiguity and over-segmentation issues, identifying all the frames in long untrimmed videos is still challenging. To address these problems, we present the Efficient Two-Step Network (ETSN) with two components. The first step of ETSN is Efficient Temporal Series Pyramid Networks (ETSPNet) that capture both local and global frame-level features and provide accurate predictions of segmentation

Accurately positioning and segmenting thymoma from computed tomography (CT) images is of great importance for an image-driven thymoma analysis. In clinical practice, the diagnosis and segmentation of thymomas for radiologists are time-consuming and inefficient tasks. Thus, it is necessary to develop a method to accurately and efficiently realize automatic segmentation of thymoma. Here, a dense skip