Explore the vast range of titles available.

Spiking neural networks (SNNs) take inspiration from the brain to enable energy-efficient computations. Since the advent of Transformers, SNNs have struggled to compete with artificial networks on modern sequential tasks, as they inherit limitations from recurrent neural networks (RNNs), with the added challenge of training with non-differentiable binary spiking activations. However, a recent renewed interest in efficient alternatives to Transformers has given rise to state-of-the-art recurrent architectures named state space models (SSMs). This work systematically investigates, for the first time, the intersection of state-of-the-art SSMs with SNNs for long-range sequence modelling. Results suggest that SSM-based SNNs can outperform the Transformer on all tasks of a well-established long-range sequence modelling benchmark. It is also shown that SSM-based SNNs can outperform current state-of-the-art SNNs with fewer parameters on sequential image classification. Finally, a novel feature mixing layer is introduced, improving SNN accuracy while challenging assumptions about the role of binary activations in SNNs. This work paves the way for deploying powerful SSM-based architectures, such as large language models, to neuromorphic hardware for energy-efficient long-range sequence modelling.

Computer-aided medical image segmentation has been applied widely in diagnosis and treatment to obtain clinically useful information of shapes and volumes of target organs and tissues. In the past several years, convolutional neural network (CNN)-based methods (e.g., U-Net) have dominated this area, but still suffered from inadequate long-range information capturing. Hence, recent work presented computer vision Transformer variants for medical image segmentation tasks and obtained promising performances. Such Transformers modeled long-range dependency by computing pair-wise patch relations. However, they incurred prohibitive computational costs, especially on 3D medical images (e.g., CT and MRI). In this paper, we propose a new method called Dilated Transformer, which conducts self-attention alternately in local and global scopes for pair-wise patch relations capturing. Inspired by dilated convolution kernels, we conduct the global self-attention in a dilated manner, enlarging receptive fields without increasing the patches involved and thus reducing computational costs. Based on this design of Dilated Transformer, we construct a U-shaped encoder–decoder hierarchical architecture called D-Former for 3D medical image segmentation. Experiments on the Synapse and ACDC datasets show that our D-Former model, trained from scratch, outperforms various competitive CNN-based or Transformer-based segmentation models at a low computational cost without time-consuming per-training process.

Six-dimensional object pose estimation is a fundamental problem in the field of computer vision. Recently, category-level object pose estimation methods based on 3D-GC have made significant breakthroughs due to advancements in 3D-GC. However, current methods often fail to capture long-range dependencies, which are crucial for modeling complex and occluded object shapes. Additionally, discerning detailed differences between different objects is essential. Some existing methods utilize self-attention mechanisms or Transformer encoder–decoder structures to address the lack of long-range dependencies, but they only focus on first-order information of features, failing to explore more complex information and neglecting detailed differences between objects. In this paper, we propose SAPENet, which follows the 3D-GC architecture but replaces the 3D-GC in the encoder part with HS-layer to extract features and incorporates statistical attention to compute higher-order statistical information. Additionally, three sub-modules are designed for pose regression, point cloud reconstruction, and bounding box voting. The pose regression module also integrates statistical attention to leverage higher-order statistical information for modeling geometric relationships and aiding regression. Experiments demonstrate that our method achieves outstanding performance, attaining an mAP of 49.5 on the 5°2 cm metric, which is 3.4 higher than the baseline model. Our method achieves state-of-the-art (SOTA) performance on the REAL275 dataset.

Long-range dependencies and contextual relationships in videos were captured by using Convolutional Neural Networks (CNNs) in past. Recently the use of Transformers is started for capturing the long-range dependencies and contextual relationships in videos. Transformers have made revolutionary impacts in Natural Language Processing (NLP) area and started making significant contributions in Computer Vision problems. So, it was required to perform the review of different Transformer Architectures in Computer Vision which will help to use them for different applications in Computer Vision. This paper provides a comprehensive review of the Transformer Architectures in Computer Vision, providing a detailed view about their evolution from Vision Transformers (ViTs) to more advanced variants of transformers like Swin Transformer, Transformer-XL, and Hybrid CNN-Transformer models. We have tried to make the study of the advantages of the Transformers over the traditional Convolutional Neural Networks (CNNs), their applications for Object Detection, Image Classification, Video Analysis, and their computational challenges. Finally, we discuss the future research directions, including the self-attention mechanisms, multi-modal learning, and lightweight architectures for Edge Computing.

The ability to assess and manage corporate credit risk enables financial institutions and investors to mitigate risk, enhance the precision of their decision-making, and adapt their strategies in a prompt and effective manner. The growing quantity of data and the increasing complexity of indicators have rendered traditional machine learning methods ineffective in enhancing the accuracy of credit risk assessment. Consequently, academics have begun to explore the potential of models based on deep learning. In this paper, we apply the concept of combining Transformer and CNN to the financial field, building on the traditional CNN-Transformer model’s capacity to effectively process local features, perform parallel processing, and handle long-distance dependencies. To enhance the model’s ability to capture financial data over extended periods and address the challenge of high-dimensional financial data, we propose a novel hybrid model, TCN-DilateFormer. This integration improves the accuracy of corporate credit risk assessment. The empirical study demonstrates that the model exhibits superior prediction accuracy compared to traditional machine learning assessment models, thereby offering a novel and efficacious tool for corporate credit risk assessment.

Remote sensing images are characterized by high complexity, significant scale variations, and abundant details, which present challenges for existing deep learning-based super-resolution reconstruction methods. These algorithms often exhibit limited convolutional receptive fields and thus struggle to establish global contextual information, which can lead to an inadequate utilization of both global and local details and limited generalization capabilities. To address these issues, this study introduces a novel multi-branch residual hybrid attention block (MBRHAB). This innovative approach is part of a proposed super-resolution reconstruction model for remote sensing data, which incorporates various attention mechanisms to enhance performance. First, the model employs window-based multi-head self-attention to model long-range dependencies in images. A multi-branch convolution module (MBCM) is then constructed to enhance the convolutional receptive field for improved representation of global information. Convolutional attention is subsequently combined across channels and spatial dimensions to strengthen associations between different features and areas containing crucial details, thereby augmenting local semantic information. Finally, the model adopts a parallel design to enhance computational efficiency. Generalization performance was assessed using a cross-dataset approach involving two training datasets (NWPU-RESISC45 and PatternNet) and a third test dataset (UCMerced-LandUse). Experimental results confirmed that the proposed method surpassed the existing super-resolution algorithms, including Bicubic interpolation, SRCNN, ESRGAN, Real-ESRGAN, IRN, and DSSR in the metrics of PSNR and SSIM across various magnifications scales.

The effectiveness of modeling contextual information has been empirically shown in numerous computer vision tasks. In this paper, we propose a simple yet efficient augmented fully convolutional network (AugFCN) by aggregating content- and position-based object contexts for semantic segmentation. Specifically, motivated because each deep feature map is a global, class-wise representation of the input, we first propose an augmented nonlocal interaction (AugNI) to aggregate the global content-based contexts through all feature map interactions. Compared to classical position-wise approaches, AugNI is more efficient. Moreover, to eliminate permutation equivariance and maintain translation equivariance, a learnable, relative position embedding branch is then supportably installed in AugNI to capture the global position-based contexts. AugFCN is built on a fully convolutional network as the backbone by deploying AugNI before the segmentation head network. Experimental results on two challenging benchmarks verify that AugFCN can achieve a competitive 45.38% mIoU (standard mean intersection over union) and 81.9% mIoU on the ADE20K val set and Cityscapes test set, respectively, with little computational overhead. Additionally, the results of the joint implementation of AugNI and existing context modeling schemes show that AugFCN leads to continuous segmentation improvements in state-of-the-art context modeling. We finally achieve a top performance of 45.43% mIoU on the ADE20K val set and 83.0% mIoU on the Cityscapes test set.

We address the point cloud semantic segmentation problem through modeling long-range dependencies based on the self-attention mechanism. Existing semantic segmentation models generally focus on local feature aggregation. By comparison, we propose a point attention network (PA-Net) to selectively extract local features with long-range dependencies. We specially devise two complementary attention modules for the point cloud semantic segmentation task. The attention modules adaptively integrate the semantic inter-dependencies with long-range dependencies. Our point attention module adaptively integrates local features of the last layer of the encoder with a weighted sum of the long-range dependency features. Regardless of the distance of similar features, they are all correlated with each other. Meanwhile, the feature attention module adaptively integrates inter-dependent feature maps among all local features in the last layer of the encoder. Extensive results prove that our two attention modules together improve the performance of semantic segmentation on point clouds. We achieve better semantic segmentation performance on two benchmark point cloud datasets (i.e., S3DIS and ScanNet). Particularly, the IoU on 11 semantic categories of S3DIS is significantly boosted.

In the realm of daily human interactions, a rich tapestry of behaviors and actions is observed, encompassing a wealth of informative cues. In the era of burgeoning big data, extensive repositories of images and videos have risen to prominence as the primary conduits for disseminating information. Grasping the intricacies of human behaviors depicted within these multimedia contexts has evolved into a pivotal quandary within the domain of computer vision. The technology of behavior recognition finds its practical application across domains such as human-computer interaction, intelligent surveillance, and anomaly detection, exhibiting a robust blend of pragmatic utility and scholarly significance. The present study introduces an innovative human body behavior recognition framework anchored in skeleton sequences and multi-stream fused spatiotemporal graph convolutional networks. Developed upon the foundation of graph convolutional networks, this method encompasses three pivotal refinements tailored to ameliorate extant challenges. First and foremost, in response to the complex task of capturing distant interdependencies among nodes within graph convolutional networks, we incorporate a spatial attention module. This module adeptly encapsulates long-term node interdependencies via precision-laden positional information, thus engendering interconnections that span diverse temporal and spatial contexts. Subsequently, to elevate the discernment of channel information within the network and to optimize the allocation of attention across distinct channels, we introduce a channel attention mechanism. This augmentation fortifies the discernment of motion-related features. Lastly, confronting the lacuna of information gaps prevalent within single-stream data, we deploy a multi-stream fusion methodology to fortify model outputs, ultimately fostering more precise prognostications concerning action classifications. Empirical results bear testament to the efficacy of the proposed multi-stream fused spatiotemporal graph convolutional network paradigm for skeleton-centric behavior recognition, evincing a pinnacle recognition accuracy of 96.0% on the expansive NTU-RGB+D skeleton dataset, alongside a zenithal accuracy of 37.3% on the Kinetics-Skeleton dataset—emanating from RGB data and furthered through pose estimation.

This paper proposes a physics-constrained 3D Swin Transformer (ST) for gravity inversion. By leveraging the self-attention mechanism in 3D ST, the method effectively models global dependencies within gravity data, enabling the network to reweight features globally and focus on critical anomalous regions. Additionally, prior gradient information is integrated into the loss function, and a hierarchical weight allocation strategy is adopted to guide the model in learning boundary information of density structures and deep-seated features more effectively. Synthetic experiments demonstrate that the proposed method achieves lower model errors, better boundary alignment, and higher inversion accuracy. The approach is further validated using gravity anomaly observations from the Gonghe Basin in Qinghai, yielding reliable and precise inversion results.