Most of manifold learning based feature extraction methods are two-step methods, which first construct a weighted neighborhood graph and then use the pre-constructed graph to perform subspace learning. As a result, these methods fail to use the underlying correlation structure of data to learn an adaptive graph to preciously characterize the similarity relationship between samples. To address this

Accurate boundary segmentation in medical images is significant yet challenging due to large variation of shape, size and appearance within intra- and inter- samples. In this paper, we present a novel deep model termed as Crossover-Net for robust segmentation in medical images. The proposed model is inspired by an interesting observation – the features learned from horizontal and vertical directions

Recent years have witnessed a trend that uses image representation models, including sparse representation (SR), low-rank representation (LRR) and their variants for multi-focus image fusion. Despite the thrilling preliminary results, existing methods conduct the fusion patch by patch, leading to insufficient consideration of the spatial consistency among the image patches within a local region or

In this paper, we address one specific video retrieval problem in terms of human face. Given one query in forms of either a frame or a sequence from a person, we search the database and return the most relevant face videos, i.e., ones have the same class label with the query. Such problem is very challenging due to the large intra-class variations and the high request on the efficiency of video representations

Hyper-Spectral Image (HSI) classification is an important task because of its wide range of applications. With the remarkable success from the Convolutional Neural Network (CNN), the performance of HSI classification has been significantly improved. However, two main challenges remained. One is that the samples of HSI have dramatic intra-class diversity and inter-class similarity, and the conventional

Comprehensive and accurate eye modelling is crucial to a variety of applications, including human-computer interaction, assistive technologies, and medical diagnosis. However, most studies focus on the localization of one or two components of eyes, such as pupil or iris, lacking a comprehensive eye model. We propose to model an eye image by a set of parametric curves. The set of curves are plotted

Graph representation learning is of paramount importance for a variety of graph analytical tasks, ranging from node classification to community detection. Recently, graph convolutional networks (GCNs) have been successfully applied for graph representation learning. These GCNs generate node representation by aggregating features from the neighborhoods, which follows the “neighborhood aggregation” scheme

The emergence of large-scale data demands new regression models with multi-dimensional coefficient arrays, known as tensor regression models. The recently proposed tensor ring decomposition has interesting properties of enhanced representation and compression capability, cyclic permutation invariance and balanced tensor ring rank, which may lead to efficient computation and fewer parameters in regression

Top-Hat transformation is an essential technology in the field of infrared small target detection. Many modified Top-Hat transformation methods have been proposed based on the different structure of structural elements. However, these methods are still hard to handle the dim targets and complex background. It can be summarized as two reasons, one is that the structural elements cannot suppress the

Unmanned aerial vehicles (UAVs) have been used in a wide range of applications and become an increasingly important radar target. To better model radar data and to tackle the curse of dimensionality, a three-step classification framework is proposed for UAV detection. First we propose to utilize the greedy subspace clustering to handle potential outliers and the complex sample distribution of radar

Video captioning aims at translating from a sequence of video frames into a sequence of words with the encoder-decoder framework. Hence, it is critical to align these two different sequences. Most existing methods exploit soft-attention (temporal attention) mechanism to align target words with corresponding frames, where the relevance of them merely depends on the previously generated words (i.e.,

Cancelable biometrics is an important biometric template protection technique. However, many existing cancelable fingerprint templates suffer post-transformation performance deterioration and the attacks via record multiplicity (ARM). In this paper, we design alignment-free cancelable fingerprint templates with dual protection, which is composed of the window-shift-XOR model and the partial discrete

Obtaining an optimal tradeoff between accuracy and efficiency in ellipse detection is a significant challenge. In this paper, we propose a fast, high-precision ellipse detection method that utilizes arc selection and grouping strategies to significantly reduce the computation amount. A fast corner detection algorithm is also proposed. In the proposed method, to generate ellipse candidates comprehensively

Due to the difficulty in acquiring massive task-specific occluded images, the classification of occluded images with deep convolutional neural networks (CNNs) remains highly challenging. To alleviate the dependency on large-scale occluded image datasets, we propose a novel approach to improve the classification accuracy of occluded images by fine-tuning the pre-trained models with a set of augmented

As for a biometric key, key management and biometric data security are both important. Existing bio-key generation methods are usually based on the biometric templates or features directly, it may expose user’s biometric data and will further make the biometric data permanently unusable for his secure identification recognitions. In this paper, a fingerprint bio-key generation approach using the feature

Cross-modal hashing similarity retrieval plays dual roles across various applications including search engines and autopilot systems. More generally, these methods also known to reduce the computation and memory storage in a training scheme. The key limitation of current methods are that: (i) they relax the discrete constrains to solve the optimization problem which may defeat the model purpose, (ii)

Subspace clustering methods have been extensively studied in recent years. For 2-dimensional (2D) data, existing subspace clustering methods usually convert 2D examples to vectors, which severely damages inherent structural information and relationships of the original data. In this paper, we propose a novel subspace clustering method, named KTRR, for 2D data. The KTRR provides us with a way to learn

Cross-modal retrieval aims at retrieving relevant points across different modalities, such as retrieving images via texts. One key challenge of cross-modal retrieval is narrowing the heterogeneous gap across diverse modalities. To overcome this challenge, we propose a novel method termed as Cross-modal discriminant Adversarial Network (CAN). Taking bi-modal data as a showcase, CAN consists of two parallel

As data always lie on a lower-dimensional space, feature selection has become an important step in computer vision, machine learning and data mining. Due to the lack of class information, the performance of unsupervised feature selection depends on how to characterize and preserve the manifold structure among data. In this paper, we propose a novel unsupervised feature selection framework, named as

In this paper, we introduce a new evidential clustering algorithm based on finding the belief-peaks and disjoint neighborhoods, called BPDNEC. The basic idea of BPDNEC is that each cluster center has the highest possibility of becoming a cluster center among its neighborhood and neighborhoods of those cluster centers are disjoint in vector space. Such possibility is measured by the belief notion in

Low-rank representation(LRR), which is a powerful method to find the low-dimensional subspace structure embedded in high-dimensional data spaces, has been used in both unsupervised learning and semi-supervised classification. LRR aims at finding the lowest rank representation that can express each data sample as linear combination of other samples. However, this method doesn’t consider the geometrical

Salient object detection (SOD) has gained tremendous attention in the field of computer vision. Multi-modal SOD based on the complementary information from RGB images and depth maps has shown remarkable success, making RGB-D saliency detection an active research topic. In this paper, we propose a novel multi-modal enhancement and fusion network (EF-Net) for effective RGB-D saliency detection. Specifically

The deep learning feature is the best for face recognition nowadays, but its performance exhibits unsatisfactorily under severe illumination variations. The main reason is that the deep learning feature was trained by the internet face images with variations of large pose/expression and slight/moderate illumination, which cannot well tackle severe illumination variations. Inspired by the fact that

Data reduction is becoming increasingly relevant due to the enormous amounts of data that are constantly being produced in many fields of research. Instance selection is one of the most widely used methods for this task. At the same time, most recent pattern recognition problems involve highly complex datasets with a large number of possible explanatory variables. For many reasons, this abundance of

Recent advanced trackers, consisting of discriminative classification component and dedicated bounding box estimation, have achieved improved performance in the visual tracking community. The most essential factor for the development is the utilization of different Convolutional Neural Networks (CNNs), which significantly improves the model capacity via offline trained deep feature representations

Traditional partition-based clustering algorithms, hard or fuzzy version of C-means, could not deal with high-dimensional data sets effectively as redundant features may impact the computation of distances and local spatial structures among patterns are rarely considered. High dimensionality of space gives rise to so-called concentration effect that is detrimental. In this paper, a novel locality preserving

Ensemble learning has been widely applied to both batch data classification and streaming data classification. For the latter setting, most existing ensemble systems are homogenous, which means they are generated from only one type of learning model. In contrast, by combining several types of different learning models, a heterogeneous ensemble system can achieve greater diversity among its members

History shows that the infectious disease (COVID-19) can stun the world quickly, causing massive losses to health, resulting in a profound impact on the lives of billions of people, from both a safety and an economic perspective, for controlling the COVID-19 pandemic. The best strategy is to provide early intervention to stop the spread of the disease. In general, Computer Tomography (CT) is used to

Gaussian discriminant analysis is a popular classification model, that in the precise case can produce unreliable predictions in case of high uncertainty (e.g., due to scarce or noisy data). While imprecise probability theory offers a nice theoretical framework to solve such issues, it has not been yet applied to Gaussian discriminant analysis. This work remedies this, by proposing a new Gaussian discriminant

Averaging GPS trajectories is needed in applications such as automatic generation of road network and finding representative movement patterns. We organized a challenge where participants submitted proposals to solve the averaging problem. In this paper, we review the proposals and evaluate their performance. We present a synthesis of the submitted methods and develop a new baseline composed of the

The distillation technique helps transform cumbersome neural networks into compact networks so that models can be deployed on alternative hardware devices. The main advantage of distillation-based approaches include a simple training process, supported by most off-the-shelf deep learning software and no special hardware requirements. In this paper, we propose a guideline for distilling the architecture

In recent years, Discriminative Correlation Filter (DCF) based tracking methods have achieved impressive performance in visual tracking. However, their excellent performance usually comes at the cost of sacrificing the computational speed. Furthermore, training correlation filters using high dimensional raw features may introduce the risk of severe over-fitting. To address the above issues, we propose

Recently, we have proposed a novel physically-inspired method, called the Nearest Descent (ND), which plays the role of organizing all the samples into an effective Graph, called the in-tree. Due to its effective characteristics, this in-tree proves very suitable for data clustering. Nevertheless, this in-tree-based clustering still has some non-trivial limitations in terms of robustness, capability

High-frequency oscillations (HFOs) are spontaneous electroencephalogram patterns that have been regarded as potential biomarkers of epileptic seizure onset zones (SOZs). Accurately detected HFOs are used to localize SOZs, which is crucial for the presurgical assessment. Since the visual marking of HFOs is time-consuming, a method is desirable to automatically detect HFOs for localizing SOZs in clinical

This paper presents a novel method for matching line segments between stereo images. Given the fundamental matrix, the local homography can be over determined with pairwise line segment candidates. We exploit this constraint to initialize the candidate and construct the novel homography graph. Because the constraint between the node is based on the epipolar geometry, the homography graph is invariant

Nonnegative matrix factorization (NMF) is a powerful dimension reduction method, and has received increasing attention in various practical applications. However, most traditional NMF based algorithms are sensitive to noisy data, or fail to fully utilize the limited supervised information. In this paper, a novel robust semi-supervised NMF method, namely correntropy based semi-supervised NMF (CSNMF)

Feature selection is a key step in many machine learning tasks. A majority of the existing methods of feature selection address the problem by devising some scoring function while treating the features independently, thereby overlooking their interdependencies. We leverage the scale invariance property of copula to construct a greedy, supervised feature selection algorithm that maximizes the feature

One-class anomaly detection approaches are particularly appealing for use in face presentation attack detection (PAD), especially in an unseen attack scenario, where the system is exposed to novel types of attacks. This work builds upon an anomaly-based formulation of the problem and analyses the merits of deploying client-specific information for face spoofing detection. We propose training one-class

Clustering is a powerful tool in exploratory data analysis that partitions a set of objects into clusters with the goal of maximizing the similarity of objects within each cluster. Due to the tendency of clustering algorithms to find suboptimal partitions of data, the approximation method Simulated Annealing (SA) has been used to search for near-optimal partitions. However, existing SA-based partitional

Human behaviours consist different types of motion; we show how they can be disambiguated into their components in a richer way than that currently possible. Studies on optical flow have concentrated on motion alone without the higher order components: snap, jerk and acceleration. We are the first to show how the acceleration, jerk, snap and their constituent parts can be obtained from image sequences

A view-graph is vital for both the accuracy and robustness of structure-from-motion (SfM). Conventional matrix decomposition techniques treat all edges of view-graph equally; hence, many edge outliers are produced in matching pairs with fewer feature matches. To address this problem, we propose an incremental framework for view-graph construction, where the robustness of matched pairs that have a larger

We consider the problem of segmenting an image into superpixels in the context of k-means clustering, in which we wish to decompose an image into local, homogeneous regions corresponding to the underlying objects. Our novel approach builds upon the widely used Simple Linear Iterative Clustering (SLIC), and incorporate a measure of objects’ structure based on the spectral residual of an image. Based

With recent advances in sensor technology, multivariate time series data are becoming extremely large with sophisticated but insightful inter-variable dependency patterns. Mining contrast dependency patterns in controlled experiments can help quantify the differences between control and experimental time series, however, overwhelms practitioners’ capability. Existing methods suffer from determining

For facial expression recognition, the sparseness constraints of the features or weights can improve the generalization ability of a deep network. However, the optimization of the hyper-parameters in fusing different sparseness strategies demands much computation, when the traditional gradient-based algorithms are used. In this work, an iterative framework with surrogate network is proposed for the

Online action detection aims to detect a current action from an untrimmed, streaming video, where only current and past frames are available. Recent methods for online action detection have focused on how to model discriminative representations from temporally partial information. However, they overlook the fact that the input video contains background as well as actions. To overcome this problem,

Scene text with an irregular layout is difficult to recognize. To this end, a Sequential Transformation Attention-based Network (STAN), which comprises a sequential transformation network and an attention-based recognition network, is proposed for general scene text recognition. The sequential transformation network rectifies irregular text by decomposing the task into a series of patch-wise basic

Facial Expression Recognition (FER) is a challenging yet important research topic owing to its significance with respect to its academic and commercial potentials. In this work, we propose an oriented attention pseudo-siamese network that takes advantage of global and local facial information for high accurate FER. Our network consists of two branches, a maintenance branch that consisted of several

Visual anomaly detection addresses the problem of classification or localization of regions in an image that deviate from their normal appearance. A popular approach trains an auto-encoder on anomaly-free images and performs anomaly detection by calculating the difference between the input and the reconstructed image. This approach assumes that the auto-encoder will be unable to accurately reconstruct

The acquisition of point clouds with a 3D scanner often yields large-scale, irregular, and unordered raw data, which hinders the classification of objects from these data. Some studies have introduced a method of applying the point clouds to convolutional neural networks (CNNs). This is achieved after preprocessing the volume metrics or multi-view images. However, this method has a limited resolution

Various AI functionalities such as pattern recognition and prediction can effectively be used to diagnose (recognize) and predict coronavirus disease 2019 (COVID-19) infections and propose timely response (remedial action) to minimize the spread and impact of the virus. Motivated by this, an AI system based on deep meta learning has been proposed in this research to accelerate analysis of chest X-ray

Scene graph parsing has become a new challenge in the field of image understanding and pattern recognition in recent years. It captures objects and their relationships, and provides a structured representation of the visual scene. Among the three types of high-level relationships of scene graphs, semantic relationships, which contain the global understanding of the scene, are the core and the most

Handwritten signature verification is used to verify the identity of individuals through recognizing their signatures. Adversarial examples can induce misclassification, hence posing a severe threat to signature verification. At present, a variety of adversarial example attacks have been developed for image classification, but they are not that useful for attacking signature verification due to two

Inspired by the temporal subspace clustering (TSC) method and low-rank matrix approximation constraint, a new model is proposed termed as temporal plus low-rank subspace clustering (TLRSC) by utilizing both the local and global structural information. On one hand, to solve the drawback that the nuclear norm-based constraint usually results in a suboptimal solution, we incorporate certain nonconvex

Discriminative feature learning is critical for pedestrian re-identification. Previous part-based methods mainly focus on the region of specific predefined semantics to learn local representations, ignoring the influence of posture changes, and the learning efficiency and robustness in complex scenes are poor. In this paper, a hypergraph video pedestrian re-identification method based on posture structure

3D feature description is one of the central techniques that rely on point clouds since a lot of point cloud processing techniques apply the point-to-point correspondences that are achieved via feature descriptors as input data. The feature descriptor encodes the information of the underlying surface around the feature point so as to make a local surface distinguished from another. The focus of the

In this paper an algorithm for multicamera pedestrian detection is proposed. The first stage of this work is based on the probabilistic occupancy map framework, in which the ground plane is discretized into a grid and the likelihood of pedestrian presence at each location is estimated by comparing a rectangle, of the average size of the pedestrians standing there, with the foreground silhouettes in

To meet the recent demands for automated security systems, this study proposes a novel single-template strategy that uses mean templates and local stability-weighted dynamic time warping (LS-DTW) to simultaneously improve the speed and accuracy of online signature verification. Specifically, we adopt a recent time-series averaging method, called Euclidean barycenter-based DTW barycenter averaging (EB-DBA)