Search

Corners, highly important local features of images and corner finding, play a crucial role in computer vision and image processing, such as object tracking and vehicle detection. Proposing effective and efficient corner detectors is the aim of corner detection. In this study, the authors first present a new measure of corner sharpness termed as the point‐to‐centroid distance (PCD) and then examine its behaviours, which display beneficial characteristics that help distinguish corners from non‐corners. Based on PCD behaviours, the authors propose a novel corner detector. Extensive experimental results demonstrate that the PCD technique is effective and simultaneously efficient for corner detection compared with six other contour‐based corner detectors in terms of two commonly used evaluation metrics – average repeatability and localisation error. [ABSTRACT FROM AUTHOR]

It is extremely difficult to find one app in app stores that exactly meets the needs of users with the boom in mobile applications nowadays. Although numerous app recommendation services are available, they mainly employ static data (e.g., information of installed apps) and rarely consider the dynamics of user interests. In this paper, we assume that user interest consists of two components: short-term temporal interests and long-term preferences, and we propose one general framework, namely DeepApp, to enhance app recommendation performance. In DeepApp, on the one hand, we use one linear model to characterize the stable user-app associations; on the other hand, we employ the Long Short-Term Memory (LSTM) model to capture the evolution of interests based on the usage patterns of mobile apps. Finally, the Wide &Deep model is applied to fuse the effects of these two types of interests, long-term preferences and short-term temporal interests, by learning the latent interaction between linear and nonlinear features. DeepApp was evaluated on a large-scale dataset, with 4,775,293 users and 238,206 mobile apps. DeepApp achieves a significant performance gain compared with baselines (more than 6% in terms of NDCG@6 over probability matrix factorization (PMF), neural collaborative filtering (NCF), and neural tensor factorization (NTF)). This demonstrates that the integration of dynamic user interests is beneficial for mobile app recommendations. [ABSTRACT FROM AUTHOR]

There are a large number of cameras in modern transportation system that capture numerous vehicle images continuously. Therefore, automatic analysis of these vehicle images is helpful for traffic flow management, criminal investigations and vehicle inspections. Vehicle matching, which aims to determine whether two input images depict an identical vehicle, is one of the core tasks in vehicle analysis. Recent relevant studies have focused on local feature extraction instead of global extraction, since local details can provide crucial cues to distinguish between cars. However, these methods do not select local features; that is, they do not assign weights to local features. Therefore, in this research, we systematically study the vehicle matching task, and present a novel annotation‐free local‐based deep learning method called Adaptive super‐pixel discriminative feature‐selective learning (ASDFL) to address this issue. In ASDFL, vehicle images are segmented into clusters of super‐pixels of similar size by considering the location and colour similarities of pixels without using any component‐level annotation. These super‐pixels are deemed to be the virtual components of vehicles. Moreover, a convolutional neural network is used to extract the deep features of these virtual components. Thereafter, an instance‐specific mask generation module driven by the extracted global features is enhanced to produce a mask to select the most distinctive virtual components of each vehicle image pair in the feature space. Finally, the vehicle matching task is accomplished by classifying the selected virtual component features of each imaged vehicle pair. Extensive experiments on two popular vehicle identification benchmarks demonstrate that our method is 1.57% and 0.8% more accurate than the previous baselines in a vehicle matching task on the VeRi and VehicleID datasets, respectively, which demonstrates the effectiveness of our method. [ABSTRACT FROM AUTHOR]

Although Out-Of-Distribution (OOD) detection has been extensively studied, the OOD detection under multi-label settings, which is closer to the real world, is still in its infancy. The pioneer work ignores some unique properties of multi-label images, such as the sparsity and co-occurrence of labels. Here, we empirically observe that these properties readily distinguish OOD and in-distribution data. Motivated by this observation, we propose a novel multi-label OOD detection approach named Sparse Label Co-occurrence Scoring (SLCS) to exploit the sparsity and co-occurrence information of labels. SLCS follows conventions and deems the logits outputted by the penultimate layer of the trained multi-label image classification model as the prediction confidences of a sample to categories in the training label set. A logit sparse filtering process is employed to filter out the low-confidence logits for avoiding the interference of low-confidence predictions while preserving the high-confidence logits to obtain the label sparsity. Then, the label co-occurrence pairs are counted for each sample based on its predicted categories and the label co-occurrence matrix constructed on the training set. Finally, the preserved logits are weighted by the label co-occurrence information and accumulated to produce the OOD detection score for each sample. Extensive experimental results on three well-known multi-label image datasets demonstrate the discriminating power of SLCS, which achieves greatly improved performances compared with the only multi-label OOD detection approach — JointEnergry and the state-of-the-art single-label OOD detection approaches. The performance improvements of SLCS over JointEnergy in FPR95 are 12.85%, 12.41%, and 9.50% on MS-COCO, VOC 2012, and NUS-WIDE datasets respectively. [ABSTRACT FROM AUTHOR]

As a commonly used data representation technique, Nonnegative Matrix Factorization (NMF) has received extensive attentions in the pattern recognition and machine learning communities over decades, since its working mechanism is in accordance with the way how the human brain recognizes objects. Inspired by the remarkable successes of manifold learning, more and more researchers attempt to incorporate the manifold learning into NMF for finding a compact representation,which uncovers the hidden semantics and respects the intrinsic geometric structure simultaneously. Graph regularized Nonnegative Matrix Factorization (GNMF) is one of the representative approaches in this category. The core of such approach is the graph, since a good graph can accurately reveal the relations of samples which benefits the data geometric structure depiction. In this paper, we leverage the sparse representation to construct a sparse hypergraph for better capturing the manifold structure of data, and then impose the sparse hypergraph as a regularization to the NMF framework to present a novel GNMF algorithm called Sparse Hypergraph regularized Nonnegative Matrix Factorization (SHNMF). Since the sparse hypergraph inherits the merits of both the sparse representation and the hypergraph model, SHNMF enjoys more robustness and can better exploit the high-order discriminant manifold information for data representation. We apply our work to address the image clustering issue for evaluation. The experimental results on five popular image databases show the promising performances of the proposed approach in comparison with the state-of-the-art NMF algorithms. [ABSTRACT FROM AUTHOR]

Motivated by the fact that the signals tend to have a representation biased towards their own classes, we propose a novel Sparse Representation-based Classifier (SRC) named Class Specific Sparse Representation-based Classifier (CSSRC), which incorporates the class information in the representation learning. Unlike the conventional SRC algorithms, CSSRC defines each class as a group and then impels these groups to compete for representing the test sample. To achieve such property, CSSRC imposes a L 1 -norm constraint to the classes for compulsively selecting the most relevant classes and introduces a L 2 -norm constraint to the samples belonging to the same class for making sure that all homogeneous samples can be sufficiently exploited for representation. Since CSSRC is a typical structure sparse representation issue, it can be efficiently solved by the convex optimization. Seven popular visual and audio signal databases are employed for evaluation. The results demonstrate its effectiveness in comparison with the state-of-the-art classifiers. [ABSTRACT FROM AUTHOR]

Human face contains abundant shape features. This fact motivates a lot of shape feature‐based face detection and three‐dimensional (3D) face recognition approaches. However, as far as we know, there is no prior low‐level face representation which is purely based on shape feature proposed for conventional 2D (image‐based) face recognition. In this study, the authors present a novel low‐level shape‐based face representation named 'shape primitives histogram' (SPH) for face recognition. In this approach, the face images are separated into a number of tiny shape fragments and they reduce these shape fragments to several uniform atomic shape patterns called 'shape primitives'. Then the face representation is obtained by implementing a histogram statistic of shape primitives in a local image region. To take scale information into consideration, they also produce multi‐scale SPHs (MSPHs) by concatenating the SPHs extracted from different scales. Moreover, they experimentally study the influences of each stage of SPH computation on performance, concluding that a small cell with 1/2 overlap and a fine size block with 1/2 overlap are important for good results. Four popular face databases, namely ORL, AR, YaleB and LFW‐a, are employed to evaluate SPH and MSPH. Surprisingly, such seemingly naive shape‐based face representations outperform the state‐of‐the‐art low‐level face representations. [ABSTRACT FROM AUTHOR]