Fast and Accurate Road Crack Detection Based on Adaptive Cost-Sensitive Loss Function.

Numerous detection problems in computer vision, including road crack detection, suffer from exceedingly foreground-background imbalance. Fortunately, modification of loss function appears to solve this puzzle once and for all. In this article, we propose a pixel-based adaptive weighted cross-entropy (WCE) loss in conjunction with Jaccard distance to facilitate high-quality pixel-level road crack detection. Our work profoundly demonstrates the influence of loss functions on detection outcomes and sheds light on the sophisticated consecutive improvements in the realm of crack detection. Specifically, to verify the effectiveness of the proposed loss, we conduct extensive experiments on four public databases, that is, CrackForest, AigleRN, Crack360, and BJN260. Compared to the vanilla WCE, the proposed loss significantly speeds up the training process while retaining the performance.

LLP-GAN: A GAN-Based Algorithm for Learning From Label Proportions.

Learning from label proportions (LLP) is a widespread and important learning paradigm: only the bag-level proportional information of the grouped training instances is available for the classification task, instead of the instance-level labels in the fully supervised scenario. As a result, LLP is a typical weakly supervised learning protocol and commonly exists in privacy protection circumstances due to the sensitivity in label information for real-world applications. In general, it is less laborious and more efficient to collect label proportions as the bag-level supervised information than the instance-level one. However, the hint for learning the discriminative feature representation is also limited as a less informative signal directly associated with the labels is provided, thus deteriorating the performance of the final instance-level classifier. In this article, delving into the label proportions, we bypass this weak supervision by leveraging generative adversarial networks (GANs) to derive an effective algorithm LLP-GAN. Endowed with an end-to-end structure, LLP-GAN performs approximation in the light of an adversarial learning mechanism without imposing restricted assumptions on distribution. Accordingly, the final instance-level classifier can be directly induced upon the discriminator with minor modification. Under mild assumptions, we give the explicit generative representation and prove the global optimality for LLP-GAN. In addition, compared with existing methods, our work empowers LLP solvers with desirable scalability inheriting from deep models. Extensive experiments on benchmark datasets and a real-world application demonstrate the vivid advantages of the proposed approach.

RPD-GAN: Learning to Draw Realistic Paintings with Generative Adversarial Network.

Painting style transfer is an attractive and challenging computer vision problem that aims to transfer painting styles onto natural images. Existing advanced methods tackle this problem from the perspective of Neural Style Transfer (NST) or unsupervised cross-domain image translation. For both two types of methods, attention has been focused on reproducing artistic painting styles of representative artists (e.g., Vincent Van Gogh). In this paper, instead of transferring styles of artistic paintings, we focus on automatic generation of realistic paintings, for example, making the machine draw a gouache before a still life, paint a sketch of a landscape, or draw a pen-and-ink portrait of a person, etc. Besides capturing the precise target styles, synthesis of realistic paintings is more demanding in preserving original content features and image structures, for which existing advanced methods are not sufficient to generate satisfactory results. Aimed at this problem, we propose RPD-GAN (Realistic Painting Drawing Generative Adversarial Network), an unsupervised cross-domain image translation framework for realistic painting style transfer. At the heart of our model is the decomposition of the image stylization mapping into four stages: feature encoding, feature de-stylization, feature re-stylization, and feature decoding, where the functionalities of these stages are implemented by additionally embedding a content-consistency constraint and a style-alignment constraint at feature space to the classic CycleGAN architecture. By enforcing these constraints, both the content-preserving and style-capturing capabilities of the model are enhanced, leading to higher-quality stylization results. Extensive experiments demonstrate the effectiveness and superiority of our RPD-GAN in drawing realistic paintings.

s-LWSR: Super Lightweight Super-Resolution Network.

In recent years, deep-based models have achieved great success in the field of single image super-resolution (SISR), where tremendous parameters are always needed to obtain a satisfying performance. However, the high computational complexity extremely limits its applications to some mobile devices that possess less computing and storage resources. To address this problem, in this paper, we propose a flexibly adjustable super lightweight SR network: s-LWSR. Firstly, in order to efficiently abstract features from the low resolution image, we design a high-efficient U-shape based block, where an information pool is constructed to mix multi-level information from the first half part of the pipeline. Secondly, a compression mechanism based on depth-wise separable convolution is employed to further reduce the numbers of parameters with negligible performance degradation. Thirdly, by revealing the specific role of activation in deep models, we remove several activation layers in our SR model to retain more information, thus leading to the final performance improvement. Extensive experiments show that our s-LWSR, with limited parameters and operations, can achieve similar performance compared with other cumbersome DL-SR methods.

Deep learning from label proportions with labeled samples.

Learning from label proportions (LLP), where the training data is in form of bags, and only the proportions of classes in each bag are available, has attracted wide interest in machine learning community. In general, most LLP algorithms adopt random sampling to obtain the proportional information of different categories, which correspondingly obtains some labeled samples in each bag. However, LLP training process always fails to leverage these labeled samples, which may contain essential data distribution information. To address this issue, in this paper, we propose end-to-end LLP solver based on convolutional neural networks (ConvNets), called LLP with labeled samples (LLP-LS). First, we reshape the cross entropy loss in ConvNets, so that it can combine the proportional information and labeled samples in each bag. Second, in order to comply with the training data in a bag manner, ADAM based on batch is employed to train LLP-LS. Hence, the batch size in training process is in accordance with the bag size. Compared with up-to-date methods on multi-class problem, our algorithm can obtain the state-of-the-art on several image datasets.

Learning from label proportions on high-dimensional data.

Learning from label proportions (LLP), in which the training data is in the form of bags and only the proportion of each class in each bag is available, has attracted wide interest in machine learning. However, how to solve high-dimensional LLP problem is still a challenging task. In this paper, we propose a novel algorithm called learning from label proportions based on random forests (LLP-RF), which has the advantage of dealing with high-dimensional LLP problem. First, by defining the hidden class labels inside target bags as random variables, we formulate a robust loss function based on random forests and take the corresponding proportion information into LLP-RF by penalizing the difference between the ground truth and estimated label proportion. Second, a simple but efficient alternating annealing method is employed to solve the corresponding optimization model. At last, various experiments demonstrate that our algorithm can obtain the best accuracies on high-dimensional data compared with several recently developed methods.

Adaboost-LLP: A Boosting Method for Learning With Label Proportions.

How to solve the classification problem with only label proportions has recently drawn increasing attention in the machine learning field. In this paper, we propose an ensemble learning strategy to deal with the learning problem with label proportions (LLP). In detail, we first give a loss function based on different weights for LLP, and then construct the corresponding weak classifier, at the same time, estimate its conditional probabilities by a standard logistic function. At last, by introducing the maximum likelihood estimation, we propose a new anyboost learning system for LLP (called Adaboost-LLP). Unlike traditional methods, our method does not make any restrictive assumptions on training set; at the same time, compared with alter- SVM, Adaboost-LLP exploits more extra weight information and uses multiple weak classifiers that can be solved efficiently to combine a strong classifier. All experiments show that our method outperforms the existing methods in both accuracy and training time.

Multi-vehicle detection with identity awareness using cascade Adaboost and Adaptive Kalman filter for driver assistant system.

Vision-based vehicle detection is an important issue for advanced driver assistance systems. In this paper, we presented an improved multi-vehicle detection and tracking method using cascade Adaboost and Adaptive Kalman filter(AKF) with target identity awareness. A cascade Adaboost classifier using Haar-like features was built for vehicle detection, followed by a more comprehensive verification process which could refine the vehicle hypothesis in terms of both location and dimension. In vehicle tracking, each vehicle was tracked with independent identity by an Adaptive Kalman filter in collaboration with a data association approach. The AKF adaptively adjusted the measurement and process noise covariance through on-line stochastic modelling to compensate the dynamics changes. The data association correctly assigned different detections with tracks using global nearest neighbour(GNN) algorithm while considering the local validation. During tracking, a temporal context based track management was proposed to decide whether to initiate, maintain or terminate the tracks of different objects, thus suppressing the sparse false alarms and compensating the temporary detection failures. Finally, the proposed method was tested on various challenging real roads, and the experimental results showed that the vehicle detection performance was greatly improved with higher accuracy and robustness.

Learning With Label Proportions via NPSVM.

Recently, learning from label proportions (LLPs), which seeks generalized instance-level predictors merely based on bag-level label proportions, has attracted widespread interest. However, due to its weak label scenario, LLP usually falls into a transductive learning framework accounting for an intractable combinatorial optimization issue. In this paper, we propose a brand new algorithm, called LLPs via nonparallel support vector machine (LLP-NPSVM), to facilitate this dilemma. To harness satisfactory data adaption, instead of transductive learning fashion, our scheme determined instance labels according to two nonparallel hyper-planes under the supervision of label proportion information. In a geometrical view, our approach can be interpreted as an alternative competitive method benefiting from large margin clustering. In practice, LLP-NPSVM can be efficiently addressed by applying two fast sequential minimal optimization paths iteratively. To rationally support the effectiveness of our method, finite termination and monotonic decrease of the proposed LLP-NPSVM procedure were essentially analyzed. Various experiments demonstrated our algorithm enjoys rapid convergence and robust numerical stability, along with best accuracies among several recently developed methods in most cases.

Nonsmooth Penalized Clustering via $\ell _{p}$ Regularized Sparse Regression.

Clustering has been widely used in data analysis. A majority of existing clustering approaches assume that the number of clusters is given in advance. Recently, a novel clustering framework is proposed which can automatically learn the number of clusters from training data. Based on these works, we propose a nonsmooth penalized clustering model via lp ( ) regularized sparse regression. In particular, this model is formulated as a nonsmooth nonconvex optimization, which is based on over-parameterization and utilizes an lp -norm-based regularization to control the tradeoff between the model fit and the number of clusters. We theoretically prove that the new model can guarantee the sparseness of cluster centers. To increase its practicality for practical use, we adhere to an easy-to-compute criterion and follow a strategy to narrow down the search interval of cross validation. To address the nonsmoothness and nonconvexness of the cost function, we propose a simple smoothing trust region algorithm and present its convergent and computational complexity analysis. Numerical studies on both simulated and practical data sets provide support to our theoretical results and demonstrate the advantages of our new method.

Successive overrelaxation for laplacian support vector machine.

Semisupervised learning (SSL) problem, which makes use of both a large amount of cheap unlabeled data and a few unlabeled data for training, in the last few years, has attracted amounts of attention in machine learning and data mining. Exploiting the manifold regularization (MR), Belkin et al. proposed a new semisupervised classification algorithm: Laplacian support vector machines (LapSVMs), and have shown the state-of-the-art performance in SSL field. To further improve the LapSVMs, we proposed a fast Laplacian SVM (FLapSVM) solver for classification. Compared with the standard LapSVM, our method has several improved advantages as follows: 1) FLapSVM does not need to deal with the extra matrix and burden the computations related to the variable switching, which make it more suitable for large scale problems; 2) FLapSVM's dual problem has the same elegant formulation as that of standard SVMs. This means that the kernel trick can be applied directly into the optimization model; and 3) FLapSVM can be effectively solved by successive overrelaxation technology, which converges linearly to a solution and can process very large data sets that need not reside in memory. In practice, combining the strategies of random scheduling of subproblem and two stopping conditions, the computing speed of FLapSVM is rigidly quicker to that of LapSVM and it is a valid alternative to PLapSVM.

Nonparallel support vector machines for pattern classification.

We propose a novel nonparallel classifier, called nonparallel support vector machine (NPSVM), for binary classification. Our NPSVM that is fully different from the existing nonparallel classifiers, such as the generalized eigenvalue proximal support vector machine (GEPSVM) and the twin support vector machine (TWSVM), has several incomparable advantages: 1) two primal problems are constructed implementing the structural risk minimization principle; 2) the dual problems of these two primal problems have the same advantages as that of the standard SVMs, so that the kernel trick can be applied directly, while existing TWSVMs have to construct another two primal problems for nonlinear cases based on the approximate kernel-generated surfaces, furthermore, their nonlinear problems cannot degenerate to the linear case even the linear kernel is used; 3) the dual problems have the same elegant formulation with that of standard SVMs and can certainly be solved efficiently by sequential minimization optimization algorithm, while existing GEPSVM or TWSVMs are not suitable for large scale problems; 4) it has the inherent sparseness as standard SVMs; 5) existing TWSVMs are only the special cases of the NPSVM when the parameters of which are appropriately chosen. Experimental results on lots of datasets show the effectiveness of our method in both sparseness and classification accuracy, and therefore, confirm the above conclusion further. In some sense, our NPSVM is a new starting point of nonparallel classifiers.

Twin support vector machine with Universum data.

The Universum, which is defined as the sample not belonging to either class of the classification problem of interest, has been proved to be helpful in supervised learning. In this work, we designed a new Twin Support Vector Machine with Universum (called U-TSVM), which can utilize Universum data to improve the classification performance of TSVM. Unlike U-SVM, in U-TSVM, Universum data are located in a nonparallel insensitive loss tube by using two Hinge Loss functions, which can exploit these prior knowledge embedded in Universum data more flexible. Empirical experiments demonstrate that U-TSVM can directly improve the classification accuracy of standard TSVM that use the labeled data alone and is superior to U-SVM in most cases.

Laplacian twin support vector machine for semi-supervised classification.

Semi-supervised learning has attracted a great deal of attention in machine learning and data mining. In this paper, we have proposed a novel Laplacian Twin Support Vector Machine (called Lap-TSVM) for the semi-supervised classification problem, which can exploit the geometry information of the marginal distribution embedded in unlabeled data to construct a more reasonable classifier and be a useful extension of TSVM. Furthermore, by choosing appropriate parameters, Lap-TSVM degenerates to either TSVM or TBSVM. All experiments on synthetic and real data sets show that the Lap-TSVM's classifier combined by two nonparallel hyperplanes is superior to Lap-SVM and TSVM in both classification accuracy and computation time.