Dynamic Vision Sensor (DVS) asynchronously records sparse events triggered by changes in pixel intensity, offering high temporal resolution and low latency. Existing frame-based methods process event data densely, violating its inherent sparsity and introducing computational redundancy. While asynchronous models preserve the event stream's native format, they often neglect spatial information, compromising their adaptability and efficiency. To address these limitations, we propose a Spatiotemporally Separated Sparse Network (S3Net) for efficient event stream encoding and learning. Specifically, we employ a learnable sparse encoding scheme to construct a voxel-structured representation that effectively extracts spatiotemporal relationships among event data. After that, we propose a dual-branch architecture to capture localized spatial dependencies and dynamic temporal patterns of event data. By explicitly decoupling spatial and temporal modeling, S3Net enables end-to-end asynchronous processing of variable-length event sequences, achieving both strong representational capacity and high computational efficiency. Experimental results on six event-based datasets demonstrate that S3Net achieves state-of-the-art performance. Compared to frame-based methods, it significantly reduces computational overhead and model complexity, while also outperforming existing asynchronous approaches in inference speed without compromising accuracy. Extensive experiments across six event-based datasets show that S3Net establishes new state-of-the-art performance. Our method reduces computational costs by 35% and model parameters by 27% compared to frame-based approaches, while delivering 1.58× faster inference than existing point-based methods at comparable accuracy levels.

Meta-Reinforcement Learning (Meta-RL) aims to enable agents to rapidly adapt to new tasks by leveraging prior experience from related ones. However, existing approaches typically assume consistent state and action spaces across tasks, limiting the ability of Meta-RL methods to generalize to environments with diverse morphologies, dynamics, and reward structures, common in real-world applications. To address this, we propose Mixture of Meta-Policies (MoMP), a modular and scalable Meta-RL framework designed for effective knowledge transfer across structurally heterogeneous environments. MoMP represents policies as sparse combinations of shared submodules, where each submodule is itself an attention-based policy block. During meta-training, MoMP learns to specialize and reuse these attention modules by optimizing their task-conditioned activation across multiple environments. Once trained, the shared module can be plugged into new agents to accelerate learning in novel tasks. Sparse activation enables targeted reuse of prior knowledge while mitigating interference, improving both adaptation speed and long-term retention. Experiments across diverse MuJoCo agents show that MoMP significantly outperforms strong meta-RL baselines, highlighting its effectiveness in cross-environment generalization and efficient meta-policy reuse.

Fault diagnosis aims to identify faults occurring in industrial production processes to prevent personnel injuries and economic losses. However, there are two main challenges in solving fault diagnosis, i.e.,extracting discriminative features from limited sensor dataandrecognizing new classes of faults. To fill these research gaps, we propose a zero-shot fault diagnosis framework, calledIF4FD, based on multiscale information fusion. First, we enhanced raw data from the perspectives of category knowledge, attribute knowledge, and feature knowledge. Then, by drawing on zero-shot learning (ZSL), we can transfer knowledge of trained faults to new classes of faults, enabling the classification of previously unknown faults. The multiscale informative knowledge effectively facilitates knowledge transfer and fault classification, thereby enhancing the accuracy of zero-shot fault diagnosis. Extensive experiments on two industrial fault diagnosis datasets validate the effectiveness of the proposed method, which consistently achieves superior performance compared to representative zero-shot fault diagnosis methods, general ZSL baselines, and several supervised classifiers. A case study on real industrial data from the Cranfield Multiphase Flow Facility also confirms the method’s effectiveness in practical applications.

Graph domain adaptation is a key subfield of graph transfer learning that aims to bridge domain gaps by transferring knowledge from a label-rich source graph to an unlabeled target graph. However, most existing methods assume balanced labels in the source graph, which often fails in practice and leads to biased knowledge transfer. To address this, in this paper, we propose a prototype-anchored learning and alignment framework for class-imbalanced graph domain adaptation. Specifically, we incorporate pointwise node mutual information into the graph encoder to capture high-order topological proximity and learn generalized node representations. Leveraging this, we then introduce categorical prototypes with adversarial proto-instances for prototype-anchored learning and recalibration to represent the source graph under an imbalanced class distribution. Finally, we introduce a weighted prototype contrastive adaptation strategy that aligns target pseudo-labels with source prototypes to handle class imbalance during adaptation. Extensive experiments show that our PALA outperforms the state-of-the-art methods. Our code is available at https://github.com/maxin88scu/PALA.

Event cameras are neuromorphic vision sensors that asynchronously capture changes in logarithmic brightness changes, offering significant advantages such as low latency, low power consumption, low bandwidth, and high dynamic range. While these characteristics make them ideal for high-speed scenarios, reconstructing geometrically consistent and photometrically accurate 3D representations from event data remains fundamentally challenging. Current event-based Neural Radiance Fields (NeRF) methods partially address these challenges but suffer from persistent artifacts caused by aggressive network learning in early stages and the inherent noise of event cameras. To overcome these limitations, we present SaENeRF, a novel self-supervised framework that effectively suppresses artifacts and enables 3D-consistent, dense, and photorealistic NeRF reconstruction of static scenes solely from event streams. Our approach normalizes predicted radiance variations based on accumulated event polarities, facilitating progressive and rapid learning for scene representation construction. Additionally, we introduce regularization losses specifically designed to suppress artifacts in regions where photometric changes fall below the event threshold and simultaneously enhance the light intensity difference of non-zero events, thereby improving the visual fidelity of the reconstructed scene. Extensive qualitative and quantitative experiments demonstrate that our method significantly reduces artifacts and achieves superior reconstruction quality compared to existing methods. The code is available at https://github.com/Mr-firework/SaENeRF.

Recently, prompt learning has achieved remarkable success in adapting pre-trained Vision-Language Models (VLMs) to downstream tasks such as image classification. However, its application to the downstream Image-Text Retrieval (ITR) task is more challenging. We find that the challenge lies in discriminating both fine-grained attributes and similar subcategories of the downstream data. To address this challenge, we propose Dual prompt Learning with Joint Category-Attribute Reweighting (DCAR), a novel dual-prompt learning framework to achieve precise image-text matching. The framework dynamically adjusts prompt vectors from both semantic and visual dimensions to improve the performance of CLIP on the downstream ITR task. Based on the prompt paradigm, DCAR jointly optimizes attribute and category features to enhance fine-grained representation learning. Specifically, (1) at the attribute level, it dynamically updates the weights of attribute descriptions based on text-image mutual information correlation; and (2) at the category level, it introduces negative samples from multiple perspectives with category-matching weighting to learn subcategory distinctions. To validate our method, we construct the Fine-class Described Retrieval Dataset (FDRD), which serves as a challenging benchmark for ITR in downstream data domains. It covers over 1,500 downstream fine categories and 230,000 image-caption pairs with detailed attribute annotations. Extensive experiments on FDRD demonstrate that DCAR achieves state-of-the-art performance over existing baselines. The code and data are available at https://github.com/wyf202322/DCAR.

Generalized zero-shot learning (GZSL) shows great potential for improving generalization to unseen classes in real-world scenarios. However, most GZSL methods depend on benchmark datasets with per-class attribute annotations, which creates a large semantic gap and worsens the domain shift problem in the visual-semantic space. To address these challenges, instance-level attributes offer an intuitive solution, but they require expensive manual annotation. In this paper, we propose a simple yet effective approach called per-instance attribute synthesis (PIAS) to generate diverse semantic representations for each instance. Our method first uses the Vision Transformer (ViT) model to extract visual features and then generates per-instance attributes. The patch splitting, positional embedding, and multi-head self-attention mechanisms in ViT improve the discriminability of both visual and semantic representations. Next, we define the generated attributes of class-average images as class anchor points. These anchor points are calibrated in the semantic space by minimizing the cosine similarity between the anchor points and per-class attribute annotations. Finally, we improve the diversity of generated per-instance attributes by aligning the topological structure between per-class attribute annotations and synthesized per-instance attributes with that between class-average visual features and per-instance visual features. We conduct comprehensive experiments on three challenging ZSL datasets: AWA2, CUB, and SUN. The results show that PIAS significantly outperforms state-of-the-art methods under both ZSL and GZSL settings. We further demonstrate the generalization ability of PIAS by applying it to attribute-based zero-shot image retrieval tasks.

Spiking neural networks (SNNs) are biologically plausible models known for their computational efficiency. A significant advantage of SNNs lies in the binary information transmission through spike trains, eliminating the need for multiplication operations. However, due to the spatio-temporal nature of SNNs, direct application of traditional backpropagation (BP) training still results in significant computational costs. Meanwhile, learning methods based on unsupervised synaptic plasticity provide an alternative for training SNNs but often yield suboptimal results. Thus, efficiently training high-accuracy SNNs remains a challenge. In this article, we propose a highly efficient and biologically plausible spiking time sparse feedback (STSF) learning method. This algorithm modifies synaptic weights by incorporating a neuromodulator for global supervised learning using sparse direct feedback alignment (DFA) and local homeostasis learning with vanilla spike-timing-dependent plasticity (STDP). Such neuromorphic global-local learning focuses on instantaneous synaptic activity, enabling independent and simultaneous optimization of each network layer, thereby improving biological plausibility, enhancing parallelism, and reducing storage overhead. Incorporating sparse fixed random feedback connections for global error modulation, which uses selection operations instead of multiplication operations, further improves computational efficiency. Experimental results demonstrate that the proposed algorithm markedly reduces the computational cost with significantly higher accuracy comparable to current state-of-the-art algorithms across a wide range of classification tasks. Our implementation codes are available at https://github.com/hppeace/STSF.

Many problems in science and engineering can be mathematically modeled using partial differential equations (PDEs), which are essential for fields like computational fluid dynamics (CFD), molecular dynamics, and dynamical systems. Although traditional numerical methods like the finite difference/element method are widely used, their computational inefficiency, due to the large number of iterations required, has long been a challenge. Recently, deep learning (DL) has emerged as a promising alternative for solving PDEs, offering new paradigms beyond conventional methods. Despite the growing interest in techniques like physics-informed neural networks (PINNs), a systematic review of the diverse neural network (NN) approaches for PDEs is still missing. This survey fills that gap by categorizing and reviewing the current progress of deep NNs (DNNs) for PDEs. Unlike previous reviews focused on specific methods like PINNs, we offer a broader taxonomy and analyze applications across scientific, engineering, and medical fields. We also provide a historical overview, key challenges, and future trends, aiming to serve both researchers and practitioners with insights into how DNNs can be effectively applied to solve PDEs.

Existing object detection models are mainly trained on large-scale labeled datasets. However, annotating data for novel aerial object classes is expensive since it is time-consuming and may require expert knowledge. Thus, it is desirable to study label-efficient object detection methods on aerial images. In this work, we propose a zero-shot method for aerial object detection named visual Description Regularization, or DescReg. Concretely, we identify the weak semantic-visual correlation of the aerial objects and aim to address the challenge with prior descriptions of their visual appearance. Instead of directly encoding the descriptions into class embedding space which suffers from the representation gap problem, we propose to infuse the prior inter-class visual similarity conveyed in the descriptions into the embedding learning. The infusion process is accomplished with a newly designed similarity-aware triplet loss which incorporates structured regularization on the representation space. We conduct extensive experiments with three challenging aerial object detection datasets, including DIOR, xView, and DOTA. The results demonstrate that DescReg significantly outperforms the state-of-the-art ZSD methods with complex projection designs and generative frameworks, e.g., DescReg outperforms best reported ZSD method on DIOR by 4.5 mAP on unseen classes and 8.1 in HM. We further show the generalizability of DescReg by integrating it into generative ZSD methods as well as varying the detection architecture. Codes will be released at https://github.com/zq-zang/DescReg.

When it comes to object detection tasks, YOLO stands out for its impressive speed and efficiency. Nonetheless, deploying YOLO on resource-constrained devices remains a challenge due to its substantial model size and memory requirements. The direct application of conventional binary quantization strategies to YOLO can result in significant accuracy degradation. A prevalent solution is to introduce floating-point shortcuts. However, the increased computational demand and parameter complexity associated with these shortcuts limit their practical deployment on hardware platforms for optimal acceleration. To solve this problem, we propose a binary neural network (BNN) for object detection called BOB-YOLO to achieve a balanced performance in terms of computational speed, model size, and detection accuracy. Our BOB-YOLO fully leverages module-wise latency (MWL) to supervise the latency of floating-point shortcut branches by that of 1-bit trunk branches. This supervision maximizes the information carried by the floating-point data flow in shortcuts while maintaining latency within the limits set by the 1-bit convolution branch, thereby improving parallel computational efficiency. We also introduce the Roofline Model to address these limitations by considering both computational complexity and parameter compression, ensuring high computational intensity. Additionally, we propose a performance evaluation metric Pd, which provides an intuitive description of the trade-off between speed and accuracy, aligning closely with the practical requirements of binary quantization strategies. Extensive experiments on the VOC and COCO datasets demonstrate the significant advantages of our method over state-of-the-art BNN methods.

Surface defect detection plays a crucial role in industrial equipment to ensure industrial safety. With the development of deep learning, a series of deep learning-based surface defect detection algorithms are proposed and achieved significant success. However, the application of the algorithms in real-world scenarios is restricted by computing power and memory resource, resulting in either deployment issues or performance degradation. In order to balance memory consumption and detection accuracy, we propose a novel method named ST-CA YOLOv5 for surface defect detection. Based on YOLOv5, The Swin Transformer Block (ST) is introduced to design the C3STR module, enhancing the ability to capture long-range semantic information. In the prediction head, we present the CAHead structure by utilizing the lightweight attention module, Coordinate Attention (CA), to fuse feature information. With the benefit of the ST and CA module, the model improves the detection ability for small objects and yields better overall detection performance. Extensive experiments conducted on several real-world datasets demonstrate the effectiveness and superiority of the proposed method, compared with the state-of-the-art methods in terms of detection performance.

In recent years, the signal diagnosis on devices operating under industrial environment has attracted increasing attention. Most data-driven signal diagnosis methods are based on a closed-set assumption that class sets of training and test data are the same. However, in industrial scenarios, during the running process of the device, the operating environment and condition may change over time, continuing generating data belonging to unknown classes with new characteristics and distribution. The unknown classes usually reflect new modes or faults of the device needed to be captured. They are unavailable in training phase, contradicting the closed-set assumption. Existing methods are inappropriate to this type of open-set classification, requiring to classify known classes and recognize unknown classes. To address this challenging problem, this article proposes a generic open-set signal classification method. First, we apply Fourier transform to convert the sensor signals from time domain to frequency domain, then data in the time and frequency domains are fused. Next, a variational encoder-classifier network is proposed to classify known classes and learn the distribution of feature space to extract robust latent features. Finally, based on extreme value theory and entropy, a pair of discriminators determine whether samples belong to unknown or not. The experimental results on two vibration-signal datasets from bearings and nuclear reactor demonstrate the effectiveness and superiority of our proposed open-set signal classification method, especially in practical applications.

In this article, we propose a structure-aligned generative adversarial network framework to improve zero-shot learning (ZSL) by mitigating the semantic gap, domain shift, and hubness problem. The proposed framework contains two parts, i.e., a generative adversarial network with a softmax classifier part, and a structure-aligned part. In the first part, the generative adversarial network aims at generating pseudovisual features through the guiding generator and discriminator play the minimax two-player game together. At the same time, the softmax classifier is committed to increasing the interclass distance and reducing intraclass distance. Then, the harmful effect of domain shift and hubness problems can be mitigated. In another part, we introduce a structure-aligned module where the structural consistency between visual space and semantic space is learned. By aligning the structure between visual space and semantic space, the semantic gap between them can be bridged. The performance of classification is improved when the structure-aligned visual-semantic embedding space is transferred to the unseen classes. Our framework reformulates the ZSL as a standard fully supervised classification task using the pseudovisual features of unseen classes. Extensive experiments conducted on five benchmark data sets demonstrate that the proposed framework significantly outperforms state-of-the-art methods in both conventional and generalized settings.

Abstract With the proposal and development of Generative Adversarial Networks, the great achievements in the field of image generation are made. Meanwhile, many works related to the generation of painting art have also been derived. However, due to the difficulty of data collection and the fundamental challenge from freehand expressions, the generation of traditional Chinese painting is still far from being perfect. This paper specialises in Chinese art painting generation of flowers, which is important and classic, by deep learning method. First, an unpaired flowers paintings data set containing three classic Chinese painting style: line drawing, meticulous, and ink is constructed. Then, based on the collected dataset, a Flower‐Generative Adversarial Network framework to generate multi‐style Chinese art painting of flowers is proposed. The Flower‐Generative Adversarial Network, consisting of attention‐guided generators and discriminators, transfers the style among line drawing, meticulous, and ink by an adversarial training way. Moreover, in order to solve the problem of artefact and blur in image generation by existing methods, a new loss function called Multi‐Scale Structural Similarity to force the structure preservation is introduced. Extensive experiments show that the proposed Flower‐Generative Adversarial Network framework can produce better and multi‐style Chinese art painting of flowers than existing methods.

Generating a new Chinese font from a multitude of references is an easy task, while it is quite difficult to generate it from a few references. In this paper, we investigate the problem of arbitrary Chinese font generation from a single reference and propose a deep learning based model, named One-reference Chinese Font Generation Network (OCFGNet), to automatically generate any arbitrary Chinese font from a single reference. Based on the disentangled representation learning, we separate the representations of stylized Chinese characters into style and content representations. Then we design a neural network consisting of the style encoder, the content encoder and the joint decoder for the proposed model. The style encoder extracts the style features of style references and maps them onto a continuous Variational Auto-Encoder (VAE) latent variable space while the content encoder extracts the content features of content references and maps them to the content representations. Finally, the joint decoder concatenates both representations in layer-wise to generate the character which has the style of style reference and the content of content reference. In addition, based on Generative Adversarial Network (GAN) structure, we adopt a patch-level discriminator to distinguish whether the received character is real or fake. Besides the adversarial loss, we not only adopt L1-regularized per-pix loss, but also combine a novel loss term Structural SIMilarity (SSIM) together to further drive our model to generate clear and satisfactory results. The experimental results demonstrate that the proposed model can not only extract style and content features well, but also have good performance in the generation of Chinese fonts from a single reference.

All things follow the cosmic equilibrium rule. As one of the important factors in evaluating images of aesthetic effects, the visual balance has not attracted much attention. In this study, a novel method is proposed to quantify the law of visual balance and automatically evaluate the aesthetic value of images. In the proposed method, the authors first analyse the colour composition of images using the K ‐means clustering method. Then, based on information aesthetics theory and perspective principles, a calculation method is proposed to find the visual centre of gravity of images. Finally, the aesthetic effect of the image based on visual balance is evaluated according to the positional relationship between the visual centre of gravity and the physical centre of the image. With extensive experimental results, the authors demonstrate qualitatively and quantitatively that the proposed evaluation method is basically consistent with the intuitive experience of most human beings. Moreover, experiments are also conducted on a large and diversified benchmark data set that are competitive with the current state of the art.