Visual industrial anomaly detection has evolved from one-class modeling to more challenging multi-class settings, where diverse categories and complex visual patterns must be jointly handled. Existing approaches often assume that anomalies lie far from normal samples in feature or spatial space. However, this assumption frequently fails due to two key issues: cross-class semantic confusion, where normal structures of one category are misclassified as anomalies in another, and pixel similarity failure, where anomalous regions visually blend into normal backgrounds. To address these challenges, we propose RPGAD (Region-Prompt Guided Anomaly Detection), an information-theoretic framework that models anomalies as semantic predictive instability, reflected in the joint responses of dual paths. RPGAD integrates two components: 1) DPENet (Dual-Path regional Energy evaluation Network), which compares region-level responses across normal-only and mixed paths through an entropy-guided energy formulation to generate robust region prompts; and 2) RDNet (Reverse Distillation Network), which selectively reconstructs prompted regions and employs a Prototype-Contrastive Optimal Transport (PCOT) loss to enhance inter-class separability and local feature aggregation. Experiments on five anomaly detection benchmarks - MVTecAD, VisA, BTAD, MPDD, and Real-IAD - demonstrate the effectiveness of RPGAD. At $256 \times 256$ resolution, RPGAD achieves strong overall performance, with mAD of 87.8%, 80.3%, 85.2%, 86.2%, and 77.7% on five benchmarks, and pixel-level AP and F1-max gains of up to 12.2 and 10.4 points over strong baselines. These results confirm that RPGAD provides accurate and robust multi-class anomaly detection and localization in complex visual scenarios.

Feature selection for high-risk disease diagnosis is critical for accurate risk assessment, yet existing methods fail to incorporate medical domain knowledge, potentially leading to suboptimal diagnostic decisions. To address this limitation, we propose recursive statistical invariants elimination (RSIE), a novel framework that systematically integrates medical domain knowledge into feature selection. RSIE comprises three complementary components: learning using statistical invariants (LUSI) for extracting reliable diagnostic patterns from domain knowledge, kernel alignment for establishing meaningful correlations between domain knowledge and dataset features, and recursive feature elimination (RFE) for systematic feature ranking and selection. We evaluate RSIE on 15 high-risk disease datasets spanning cardiovascular, neurological, and oncological conditions. Experimental results show that RSIE achieves an average accuracy of 88.74% across all datasets compared to 83.28% for traditional SVM-RFE, representing a 5.46% improvement. Specifically, the LUSI-RFE* variant achieves superior performance, reducing feature dimensions by 45% -74% while maintaining high classification accuracy. The experiments simultaneously demonstrate the stability of RSIE against high noise conditions, computational efficiency, and model interpretability for high-risk disease diagnosis.

Efficient and accurate detection of surface defects on trains is crucial for ensuring train safety. However, the insufficient defect samples and their diverse patterns make defect detection in complex environments highly challenging. This paper proposes a novel train surface defect detection model (ViLG) via visual-language knowledge guidance. By leveraging broad semantic knowledge of CLIP, the model compensates for the insufficient defect semantics in tiny datasets and enhances the ability to recognize unseen defects. First, we propose Visual Feature Guidance with CLIP, which enriches and enhances the global representation capabilities of backbone while preserving its self-learning ability for visual representations. This improves semantic understanding of complex scenarios and diverse defects. Second, we propose Defect Query Selector, which selects defect queries based on the semantic relevance between texts and global feature embeddings. This increases attention to potential defects and reduces missed detections. Finally, we propose Semantic Consistency Loss, which semantically aligns defect queries with defect prompts. With additional cross-modal supervision signals, it refines the semantics of defects. For real-world scenarios with normal reference images, we propose ViLG+, which effectively filters false positives using feature similarity. It further verifies that the global embeddings effectively represent the overall structure of visual scenes as well as subtle local features. Compared with other advanced methods on two train surface defect datasets and two public defect datasets, ViLG shows higher precision, recall, and average precision on unseen defects with relatively faster speed, with average improvements of 23.58, 3.23, and 6.05, and has a more balanced false positive rate and false negative rate.

Pixel-level road crack detection has always been a challenging task in intelligent transportation systems. Due to the external environments, such as weather, light, and other factors, pavement cracks often present low contrast, poor continuity, and different sizes in length and width. However, most of the existing studies pay less attention to crack data under different situations. Meanwhile, recent algorithms based on deep convolutional neural networks (DCNNs) have promoted the development of cutting-edge models for crack detection. Nevertheless, they usually focus on complex models for good performance, but ignore detection efficiency in practical applications. In this article, to address the first issue, we collected two new databases (i.e. Rain365 and Sun520) captured on rainy and sunny days respectively, which enrich the data of the open source community. For the second issue, we reconsider how to improve detection efficiency with excellent performance, and then propose our lightweight encoder-decoder architecture termed CarNet. Specifically, we introduce a novel olive-shaped structure for the encoder network, a lightweight multi-scale block and a new up-sampling method in the decoder network. Numerous experiments show that our model can better balance detection performance and efficiency compared with previous models. Especially, on the Sun520 dataset, our CarNet significantly advances the state-of-the-art performance with ODS F-score from 0.488 to 0.514. Meanwhile, it does so with an improved detection speed (104 frames per second) which is orders of magnitude faster than some recent DCNNs-based algorithms specially designed for crack detection.

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.

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.

We propose an end-to-end-trainable feature augmentation module built for image classification that extracts and exploits multi-view local features to boost model performance. Different from using global average pooling (GAP) to extract vectorized features from only the global view, we propose to sample and ensemble diverse multi-view local features to improve model robustness. To sample class-representative local features, we incorporate a simple auxiliary classifier head (comprising only one 1x1 convolutional layer) which efficiently and adaptively attends to class-discriminative local regions of feature maps via our proposed AdaCAM (Adaptive Class Activation Mapping). Extensive experiments demonstrate consistent and noticeable performance gains achieved by our multi-view feature augmentation module.

Learning from label proportions (LLP) aims at learning an instance-level classifier with label proportions in grouped training data. Existing deep learning based LLP methods utilize end-to-end pipelines to obtain the proportional loss with Kullback-Leibler divergence between the bag-level prior and posterior class distributions. However, the unconstrained optimization on this objective can hardly reach a solution in accordance with the given proportions. Besides, concerning the probabilistic classifier, this strategy unavoidably results in high-entropy conditional class distributions at the instance level. These issues further degrade the performance of the instance-level classification. In this paper, we regard these problems as noisy pseudo labeling, and instead impose the strict proportion consistency on the classifier with a constrained optimization as a continuous training stage for existing LLP classifiers. In addition, we introduce the mixup strategy and symmetric cross-entropy to further reduce the label noise. Our framework is model-agnostic, and demonstrates compelling performance improvement in extensive experiments, when incorporated into other deep LLP models as a post-hoc phase.

Natural image matting is an important problem that widely applied in computer vision and graphics. Recent deep learning matting approaches have made an impressive process in both accuracy and efficiency. However, there are still two fundamental problems remain largely unsolved: 1) accurately separating an object from the image with similar foreground and background color or lots of details; 2) exactly extracting an object with fine structures from complex background. In this paper, we propose an attention transfer network (ATNet) to overcome these challenges. Specifically, we firstly design a feature attention block to effectively distinguish the foreground object from the color-similar regions by activating foreground-related features as well as suppressing others. Then, we introduce a scale transfer block to magnify the feature maps without adding extra information. By integrating the above blocks into an attention transfer module, we effectively reduce the artificial content in results and decrease the computational complexity. Besides, we use a perceptual loss to measure the difference between the feature representations of the predictions and the ground-truths. It can further capture the high-frequency details of the image, and consequently, optimize the fine structures of the object. Extensive experiments on two publicly common datasets (i.e., Composition-1k matting dataset, and www.alphamatting.com dataset) show that the proposed ATNet obtains significant improvements over the previous methods. The source code and compiled models have been made publicly available at https://github.com/ailsaim/ATNet.

This paper develops a multi-task learning framework that attempts to incorporate the image structure knowledge to assist image inpainting, which is not well explored in previous works. The primary idea is to train a shared generator to simultaneously complete the corrupted image and corresponding structures — edge and gradient, thus implicitly encouraging the generator to exploit relevant structure knowledge while inpainting. In the meantime, we also introduce a structure embedding scheme to explicitly embed the learned structure features into the inpainting process, thus to provide possible preconditions for image completion. Specifically, a novel pyramid structure loss is proposed to supervise structure learning and embedding. Moreover, an attention mechanism is developed to further exploit the recurrent structures and patterns in the image to refine the generated structures and contents. Through multi-task learning, structure embedding besides with attention, our framework takes advantage of the structure knowledge and outperforms several state-of-the-art methods on benchmark datasets quantitatively and qualitatively.

In this article, we present a general framework for low-level vision tasks including image compression artifacts reduction and image denoising. Under this framework, a novel concatenated attention neural network (CANet) is specifically designed for image restoration. The main contributions of this article are as follows: First, we establish a concise network with a recursive topology by abandoning most complex artificially designed topology of the network. Second, we did not use the down-sampling mechanism in our network and a lightweight attention mechanism is adopted for this. Third, we rethink the feature fusion mechanism in the field of image restoration and improve this by applying concise but effective concatenation and feature selection mechanism which promotes further extraction of more cleaner features in images. Lastly, we demonstrate that CANet achieves better results than previous state-of-the-art approaches with sufficient experiments in compression artifacts removing and image denoising.

To make better use of the cloud computing technology, and to overcome the computing and storage requirements which increase rapidly with the number of training samples, in this paper, a new parallel algorithm is proposed-Parallel Regularized Multiple-Criteria Linear Programming (PRMCLP) algorithm-The RMCLP model is converted into a unconstrained optimization problem, and then, in the parallel version, it is split into several tasks, where each part is mapped and computed on a separate processor. This approach enables us to obtain efficiently the final optimization solution of the whole classification problem. At last, we apply this algorithm to Medical data classification. All experiments show that our method and approach greatly increases the training speed of RMCLP in the parallel case.

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.

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.

In this paper, we leverage generative adversarial networks (GANs) to derive an effective algorithm LLP-GAN for learning from label proportions (LLP), where only the bag-level proportional information in labels is available. Endowed with end-to-end structure, LLP-GAN performs approximation in the light of an adversarial learning mechanism, without imposing restricted assumptions on distribution. Accordingly, we can directly induce the final instance-level classifier upon the discriminator. Under mild assumptions, we give the explicit generative representation and prove the global optimality for LLP-GAN. Additionally, compared with existing methods, our work empowers LLP solver with capable scalability inheriting from deep models. Several experiments on benchmark datasets demonstrate vivid advantages of the proposed approach.

Learning from label proportions (LLP) is a new kind of learning problem which has attracted wide interest in the field of machine learning. Different from the well-known supervised learning, the training data of LLP is in form of bags and only the proportion of each class in each bag is available. Actually, many modern applications can be abstracted to this problem such as modeling voting behaviors and spam filtering. In this paper, we propose an end-to-end LLP model based on convolutional neural network called IDLLP, which employs the the idea of inverting a classifier calibration process to learn a classifier from bag probabilities. Firstly, convolutional neural network regression is used to estimate the values obtained by inverting the probability of each bag. Secondly, stochastic gradient descent based on batch is adapt to train the model, where the batch size depends on the bag size. At last, experiments demonstrate that our algorithm can obtain the best accuracies on image data compared with several recently developed methods.

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.

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(0p-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.

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.

Cracks are a growing threat to road conditions and have drawn much attention to the construction of intelligent transportation systems. However, as the key part of an intelligent transportation system, automatic road crack detection has been challenged because of the intense inhomogeneity along the cracks, the topology complexity of cracks, the inference of noises with similar texture to the cracks, and so on. In this paper, we propose CrackForest, a novel road crack detection framework based on random structured forests, to address these issues. Our contributions are shown as follows: 1) apply the integral channel features to redefine the tokens that constitute a crack and get better representation of the cracks with intensity inhomogeneity; 2) introduce random structured forests to generate a high-performance crack detector, which can identify arbitrarily complex cracks; and 3) propose a new crack descriptor to characterize cracks and discern them from noises effectively. In addition, our method is faster and easier to parallel. Experimental results prove the state-of-the-art detection precision of CrackForest compared with competing methods.

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.

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.