Search

Recently, crowd counting via estimating a density map has been widely studied. However, it still has a variety of issues to overcome, such as large-scale variation of population, complex background noise, perspective distortion, etc. The large-scale variation of heads will restrict the performance of crowd counting approaches, and the complex background noise will result in the background, such as leaf and mesh, being incorrectly recognized as heads. To maintain large-scale variation and generate a high-quality estimated density map, we propose a novel multi-scale fusion scale-aware attention network called multi-scale and gated spatial attention network (MGSNet). In MGSNet, the first 10 layers of VGG16 with Batch Normalization (BN) are utilized as backbone. Then, two branches, i.e., a large-scale branch and a scale–aware attention branch, are followed. The large-scale branch is used to overcome the large-scale variation of heads in crowd images, in which a Scale Information Aggregation Block (SIAB) is employed to extract multi-scale features by utilizing dilated convolution with different receptive fields. The scale-aware attention branch is used to address complex background noise in crowd scenes, in which a Gated Spatial Attention Block (GSAB) inspired by the Long Short-term Memory Networks (LSTM) is employed to fuse the previous information with different scales and retain the appropriate scale information of crowds. We demonstrate our proposed method on the ShanghaiTech (Part AB), UCF-CC-50 and UCF-QNRF datasets. The experimental results show its effectiveness over the state-of-the-art. [ABSTRACT FROM AUTHOR]

Fine-grained information has been proved helpful for person re-identification, and multi-head attention mechanism offers a feasible solution for this. However, we observe severe redundancy among the multiple branches, which might make the learned representation over-emphasize certain discriminative regions and correspondingly ignore other potentially informative regions. Therefore, we tackle this issue by two aspects yielding the so-called Complementation-Reinforced Attention Network (CRAN). One is the redundancy among branches, and we propose to impose complementing constraints among multiple attention heads. The constraints are two-fold: on the one hand, it encourages each branch to attend to complementary attention regions; on the other hand, it enforces orthogonality among the learned features of different regions in the embedding space. The other is the redundancy among query positions for each attention head. So we simplify the attention block by sparsifying the query positions. Besides, in order to achieve efficient retrieval, we propose an adaptive feature fusion method for dimensional reduction. Compared with the commonly used feature ensemble, our method effectively reduces the dimensionality while keeping the discriminative ability. We demonstrate the effectiveness of our method on MSMT17, Market-1501, DukeMTMC-reID, and CUHK03 datasets. [ABSTRACT FROM AUTHOR]

We present an effective semi-supervised learning algorithm for single image dehazing. The proposed algorithm applies a deep Convolutional Neural Network (CNN) containing a supervised learning branch and an unsupervised learning branch. In the supervised branch, the deep neural network is constrained by the supervised loss functions, which are mean squared, perceptual, and adversarial losses. In the unsupervised branch, we exploit the properties of clean images via sparsity of dark channel and gradient priors to constrain the network. We train the proposed network on both the synthetic data and real-world images in an end-to-end manner. Our analysis shows that the proposed semi-supervised learning algorithm is not limited to synthetic training datasets and can be generalized well to real-world images. Extensive experimental results demonstrate that the proposed algorithm performs favorably against the state-of-the-art single image dehazing algorithms on both benchmark datasets and real-world images. [ABSTRACT FROM AUTHOR]

Highlights • Motion blur kernel size is a significant input parameter of deblurring algorithms. • A regression CNN is proposed to estimate the kernel size of a blurry image. • The kernel size for training is labeled in a relative way. • The kernel size to be estimated can be arbitrary. • The estimated kernel size helps existing deblurring algorithms yielding satisfactory results. Abstract Deblurring is to restore a latent clear image as well as to estimate an underlying blur kernel from a single blurry image. Motion blur kernel size is a significant input parameter of existing deblurring algorithms. Setting the size manually, which is a tedious trial-and-error process, can hardly obtain a satisfactory result directly and effectively. In this paper, we formulate the kernel size estimation as a regression problem and construct a convolutional neural network (CNN) to resolve it. After training the CNN by a large number of simulated blurry images with labels, which are produced by a relative way, we can estimate the kernel sizes of a given blurry image without size limitation. Both quantitative and qualitative results of experiments show that our method can not only estimate the motion blur kernel size accurately but also help to yield better pictures with estimated kernel sizes as input parameters. [ABSTRACT FROM AUTHOR]

Person reidentification (re-id) aims to match a specific person across nonoverlapping cameras, which is an important but challenging task in video surveillance. Conventional methods mainly focus either on feature constructing or metric learning. Recently, some deep learning-based methods have been proposed to learn image features and similarity measures jointly. However, current deep models for person re-id are usually trained with either pairwise loss, where the number of negative pairs greatly outnumbering that of positive pairs may lead the training model to be biased toward negative pairs or constant margin hinge loss, without considering the fact that hard negative samples should be paid more attention in the training stage. In this paper, we propose to learn deep representations with an adaptive margin listwise loss. First, ranking lists instead of image pairs are used as training samples, in this way, the problem of data imbalance is relaxed. Second, by introducing an adaptive margin parameter in the listwise loss function, it can assign larger margins to harder negative samples, which can be interpreted as an implementation of the automatic hard negative mining strategy. To gain robustness against changes in poses and part occlusions, our architecture combines four convolutional neural networks, each of which embeds images from different scales or different body parts. The final combined model performs much better than each single model. The experimental results show that our approach achieves very promising results on the challenging CUHK03, CUHK01, and VIPeR data sets. [ABSTRACT FROM PUBLISHER]

Abstract: In this paper, we present a novel method to detect license plates in video sequences automatically. The framework mainly integrates the cascade detectors method and the Tracking Learning Detection (TLD) algorithm. The cascade detectors are used to detect license plates, and the TLD algorithm is adopted to track the license plate regions. The license plates in the first frame image are detected by the cascade detectors to build the original tracking list, the tracking results and the detection results in following frames will be compared, and the newly appearing license plate information will be added to the tracking list. Meanwhile, the tracking results existing in the current tracking list would be replaced by the corresponding detection results with higher degree of confidence. We demonstrate the effectiveness of our algorithm for license plate detection task on a road intersection dataset, and the experimental evaluation shows the detection performance has been greatly improved by synthetically using the detection and tracking strategy. [Copyright &y& Elsevier]

Abstract: In the object recognition process of the remote sensing image, as the object data is influenced by the imaging scale, intensity and the shape of the object, the distribution of the object data is very different with that of known training data which leads to the low reliability of the object recognition. Aimed at the problem, an object recognition method based on transfer learning framework for the remote sensing image is proposed in this paper. The feature vectors of the object data are extracted by an improved LBP firstly, and then the transfer learning is used to find the common parameters among the feature spaces of the object data with the different distributions. The transfer learning method can transfer knowledge from the old object data to the new object data and improve the performance of the object recognition. According to the experiments in the satellite remote sensing images, it shows that the accuracy of object recognition has been greatly improved by our proposed method compared with the other classical methods. [Copyright &y& Elsevier]

Abstract: Despite of its great success, two key problems are still unresolved for AdaBoost algorithms: how to select the most discriminative weak learners and how to optimally combine them. In this paper, a new AdaBoost algorithm is proposed to make improvement in the two aspects. First, we select the most discriminative weak learners by minimizing a novel distance related criterion, i.e., error-degree-weighted training error metric (ETEM) together with generalization capability metric (GCM), rather than training error rate only. Second, after getting the coefficients that are set empirically, we combine the weak learners optimally by tuning the coefficients using kernel-based perceptron. Experiments with synthetic and real scene data sets show our algorithm outperforms conventional AdaBoost. [Copyright &y& Elsevier]

Video synopsis is an intelligent condensation approach to solve fast video browsing and retrieval for surveillance cameras. However, collision caused by unsatisfied tube rearrangement in traditional methods brings uncomfortable visual effect to users and how to mitigate the collision still remains an attracting topic. Unlike conventional methods that deal with tube rearrangement by minimizing a global energy function, we propose a novel approach by formulating it as a graph coloring problem. In our approach, all the tubes are firstly mapped into the spatial domain for analyzing their potential collision relationship. The input tube set is then represented by a graph structure, where each node stands for a tube and the edge between two nodes represents the potential collision relationship. To mitigate the collision artifacts, our method finds the mapping of tubes from original video to synopsis video by L ( q )- coloring the graph, which separates tubes from their collision points. The parameter q is left tunable to make a compromise between collision artifacts and synopsis length, which can better meet users’ demand of freely adjusting the compactness of synopsis video. The shifted objects are finally composited with the background image to obtain the high-quality video synopsis. Extensive experimental results show that the proposed method can generate more compact video synopsis with less collision artifacts than the existing methods. [ABSTRACT FROM AUTHOR]

Person re-identification (re-id), aiming to search a specific person among a non-overlapping camera network, has attracted plenty of interest in recent years. This task is highly challenging, especially when there exists only single image per person in the database. In this paper, we present an algorithm for learning a Mahalanobis distance for person re-identification. Our method has two distinctive features: (1) to obtain the best separability of the training data, we first minimize the intra-class distances to the most extent by forcing intra-class distances to be zero, and (2) to promote the generalization ability of the learned metric, we then maximize the minimum margin between different classes. Inspired by the simple geometric intuition that a regular simplex maximizes its minimum side length, provided the sum of all side length is fixed, our method, called EquiDistance constrained Metric Learning (EquiDML), applies least-square regression technique to map images of the same person to the same vertex of a regular simplex, and images of different persons to different vertices of a regular simplex. Consequently, under the learned metric, images of the same class are collapsed to a single point, while images of different classes are transformed to be equidistant. This simple motivation is further formulated as a convex optimization problem, solved by the projected gradient descent method and proved to be very effective in person re-identification task. Although it is fairly simple, our method outperforms the state-of-the-art methods on CUHK01, CUHK03, Market1501 and DukeMTMC-reID datasets, and achieves very competitive performance on the widely used VIPeR dataset. [ABSTRACT FROM AUTHOR]