Parallel Features Research Articles

QoS-pro: A QoS-enhanced Transaction Processing Framework for Shared SSDs

Solid State Drives (SSDs) are widely used in data-intensive scenarios due to their high performance and decreasing cost. However, in shared environments, concurrent workloads can interfere with each other, leading to a violation of Quality of Service (QoS). While QoS mechanisms like fairness guarantees and latency constraints have been integrated into SSDs, existing transaction processing frameworks offer limited QoS guarantees and can significantly degrade overall performance in a shared environment. The reason is that the internal components of an SSD, originally designed to exploit parallelism, struggle to coordinate effectively when QoS mechanisms are applied to them. This article proposes a novel QoS -enhanced transaction pro cessing framework, called QoS-pro, which enhances QoS guarantees for concurrent workloads while maintaining high parallelism for SSDs. QoS-pro achieves this by redesigning transaction processing procedures to fully exploit the parallelism of shared SSDs and enhancing QoS-oriented transaction translation and scheduling with parallelism features in mind. In terms of fairness guarantees, QoS-pro outperforms state-of-the-art methods by achieving 96% fairness improvement and 64% maximum latency reduction. QoS-pro also shows almost no loss in throughput when compared with parallelism-oriented methods. Additionally, QoS-pro triggers the fewest Garbage Collection (GC) operations and minimally affects concurrently running workloads during GC operations.

Research on a Mongolian Text to Speech Model Based on Ghost and ILPCnet

The core challenge of speech synthesis technology is how to convert text information into an audible audio form to meet the needs of users. In recent years, the quality of speech synthesis based on end-to-end speech synthesis models has been significantly improved. However, due to the characteristics of the Mongolian language and the lack of an audio corpus, the Mongolian speech synthesis model has achieved few results, and there are still some problems with the performance and synthesis quality. First, the phoneme information of Mongolian was further improved and a Bang-based pre-training model was constructed to reduce the error rate of Mongolian phonetic synthesized words. Second, a Mongolian speech synthesis model based on Ghost and ILPCnet was proposed, named the Ghost-ILPCnet model, which was improved based on the Para-WaveNet acoustic model, replacing ordinary convolution blocks with stacked Ghost modules to generate Mongolian acoustic features in parallel and improve the speed of speech generation. At the same time, the improved vocoder ILPCnet had a high synthesis quality and low complexity compared to other vocoders. Finally, a large number of data experiments were conducted on the proposed model to verify its effectiveness. The experimental results show that the Ghost-ILPCnet model has a simple structure, fewer model generation parameters, fewer hardware requirements, and can be trained in parallel. The average subjective opinion score of its synthesized speech reached 4.48 and the real-time rate reached 0.0041. It ensures the naturalness and clarity of synthesized speech, speeds up the synthesis speed, and effectively improves the performance of the Mongolian speech synthesis model.

Real‐world image deblurring using data synthesis and feature complementary network

AbstractMany learning‐based approaches to image deblurring have received increasing attention in recent years. However, the models trained on existing synthetic datasets do not generalize well to real‐world blur, resulting in undesirable artifacts and residual blur. This work attempts to address this problem from two aspects: training data synthesis and network architecture. To narrow the domain gap between synthetic and real domains, a realistic blur synthesis pipeline to generate high‐quality blurred data is proposed. Since the blur is non‐uniform and has different scales and degrees, a parallel feature complementary module to fully exploit the local and non‐local information, which improves the feature representation and helps the network to perceive the non‐uniform blur, is developed. In addition, a spatial Fourier reconstruction block to facilitate correct detail recovery in the spatial and Fourier domains is introduced. Based on these two designs, an effective encoder–decoder network for deblurring is designed. Extensive experiments demonstrate the validity and superiority of the proposed blur synthesis method and deblurring network. In particular, the proposed deblurring network can achieve superior or comparable performance to Restormer, while saving 70% of network parameters and 53% of floating point operations (FLOPs).

CVTrack: Combined Convolutional Neural Network and Vision Transformer Fusion Model for Visual Tracking.

Most single-object trackers currently employ either a convolutional neural network (CNN) or a vision transformer as the backbone for object tracking. In CNNs, convolutional operations excel at extracting local features but struggle to capture global representations. On the other hand, vision transformers utilize cascaded self-attention modules to capture long-range feature dependencies but may overlook local feature details. To address these limitations, we propose a target-tracking algorithm called CVTrack, which leverages a parallel dual-branch backbone network combining CNN and Transformer for feature extraction and fusion. Firstly, CVTrack utilizes a parallel dual-branch feature extraction network with CNN and transformer branches to extract local and global features from the input image. Through bidirectional information interaction channels, the local features from the CNN branch and the global features from the transformer branch are able to interact and fuse information effectively. Secondly, deep cross-correlation operations and transformer-based methods are employed to fuse the template and search region features, enabling comprehensive interaction between them. Subsequently, the fused features are fed into the prediction module to accomplish the object-tracking task. Our tracker achieves state-of-the-art performance on five benchmark datasets while maintaining real-time execution speed. Finally, we conduct ablation studies to demonstrate the efficacy of each module in the parallel dual-branch feature extraction backbone network.

CS2DIPs: Unsupervised HSI Super-Resolution Using Coupled Spatial and Spectral DIPs.

In recent years, fusing high spatial resolution multispectral images (HR-MSIs) and low spatial resolution hyperspectral images (LR-HSIs) has become a widely used approach for hyperspectral image super-resolution (HSI-SR). Various unsupervised HSI-SR methods based on deep image prior (DIP) have gained wide popularity thanks to no pre-training requirement. However, DIP-based methods often demonstrate mediocre performance in extracting latent information from the data. To resolve this performance deficiency, we propose a coupled spatial and spectral deep image priors (CS2DIPs) method for the fusion of an HR-MSI and an LR-HSI into an HR-HSI. Specifically, we integrate the nonnegative matrix-vector tensor factorization (NMVTF) into the DIP framework to jointly learn the abundance tensor and spectral feature matrix. The two coupled DIPs are designed to capture essential spatial and spectral features in parallel from the observed HR-MSI and LR-HSI, respectively, which are then used to guide the generation of the abundance tensor and spectral signature matrix for the fusion of the HSI-SR by mode-3 tensor product, meanwhile taking some inherent physical constraints into account. Free from any training data, the proposed CS2DIPs can effectively capture rich spatial and spectral information. As a result, it exhibits much superior performance and convergence speed over most existing DIP-based methods. Extensive experiments are provided to demonstrate its state-of-the-art overall performance including comparison with benchmark peer methods.

Cascaded Inner-Outer Clip Retformer for Ultrasound Video Object Segmentation.

Computer-aided ultrasound (US) imaging is an important prerequisite for early clinical diagnosis and treatment. Due to the harsh ultrasound (US) image quality and the blurry tumor area, recent memory-based video object segmentation models (VOS) achieve frame-level segmentation by performing intensive similarity matching among the past frames which could inevitably result in computational redundancy. Furthermore, the current attention mechanism utilized in recent models only allocates the same attention level among whole spatial-temporal memory features without making distinctions, which may result in accuracy degradation. In this paper, we first build a larger annotated benchmark dataset for breast lesion segmentation in ultrasound videos, then we propose a lightweight clip-level VOS framework for achieving higher segmentation accuracy while maintaining the speed. The Inner-Outer Clip Retformer is proposed to extract spatialtemporal tumor features in parallel. Specifically, the proposed Outer Clip Retformer extracts the tumor movement feature from past video clips to locate the current clip tumor position, while the Inner Clip Retformer detailedly extracts current tumor features that can produce more accurate segmentation results. Then a Clip Contrastive loss function is further proposed to align the extracted tumor features along both the spatial-temporal dimensions to improve the segmentation accuracy. In addition, the Global Retentive Memory is proposed to maintain the complementary tumor features with lower computing resources which can generate coherent temporal movement features. In this way, our model can significantly improve the spatial-temporal perception ability without increasing a large number of parameters, achieving more accurate segmentation results while maintaining a faster segmentation speed. Finally, we conduct extensive experiments to evaluate our proposed model on several video object segmentation datasets, the results show that our framework outperforms state-of-theart segmentation methods.

An Interpretable Adaptive Multiscale Attention Deep Neural Network for Tabular Data.

Deep learning (DL) has been demonstrated to be a valuable tool for analyzing signals such as sounds and images, thanks to its capabilities of automatically extracting relevant patterns as well as its end-to-end training properties. When applied to tabular structured data, DL has exhibited some performance limitations compared to shallow learning techniques. This work presents a novel technique for tabular data called adaptive multiscale attention deep neural network architecture (also named excited attention). By exploiting parallel multilevel feature weighting, the adaptive multiscale attention can successfully learn the feature attention and thus achieve high levels of F1-score on seven different classification tasks (on small, medium, large, and very large datasets) and low mean absolute errors on four regression tasks of different size. In addition, adaptive multiscale attention provides four levels of explainability (i.e., comprehension of its learning process and therefore of its outcomes): 1) calculates attention weights to determine which layers are most important for given classes; 2) shows each feature's attention across all instances; 3) understands learned feature attention for each class to explore feature attention and behavior for specific classes; and 4) finds nonlinear correlations between co-behaving features to reduce dataset dimensionality and improve interpretability. These interpretability levels, in turn, allow for employing adaptive multiscale attention as a useful tool for feature ranking and feature selection.

Application of Convolutional Neural Networks for Parallel Multi-Scale Feature Extraction in Noise Image Denoising

Application of Convolutional Neural Networks for Parallel Multi-Scale Feature Extraction in Noise Image Denoising

DPNet: Dual-Path Network for Real-Time Object Detection With Lightweight Attention.

The recent advances in compressing high-accuracy convolutional neural networks (CNNs) have witnessed remarkable progress in real-time object detection. To accelerate detection speed, lightweight detectors always have few convolution layers using a single-path backbone. Single-path architecture, however, involves continuous pooling and downsampling operations, always resulting in coarse and inaccurate feature maps that are disadvantageous to locate objects. On the other hand, due to limited network capacity, recent lightweight networks are often weak in representing large-scale visual data. To address these problems, we present a dual-path network, named DPNet, with a lightweight attention scheme for real-time object detection. The dual-path architecture enables us to extract in parallel high-level semantic features and low-level object details. Although DPNet has a nearly duplicated shape with respect to single-path detectors, the computational costs and model size are not significantly increased. To enhance representation capability, a lightweight self-correlation module (LSCM) is designed to capture global interactions, with only a few computational overheads and network parameters. In the neck, LSCM is extended into a lightweight cross correlation module (LCCM), capturing mutual dependencies among neighboring scale features. We have conducted exhaustive experiments on MS COCO, Pascal VOC 2007, and ImageNet datasets. The experimental results demonstrate that DPNet achieves a state-of-the-art trade off between detection accuracy and implementation efficiency. More specifically, DPNet achieves 31.3% AP on MS COCO test-dev, 82.7% mAP on Pascal VOC 2007 test set, and 41.6% mAP on ImageNet validation set, together with nearly 2.5M model size, 1.04 GFLOPs, and 164 and 196 frames/s (FPS) FPS for [Formula: see text] input images of three datasets.

Gloss Prior Guided Visual Feature Learning for Continuous Sign Language Recognition.

Continuous sign language recognition (CSLR) is to recognize the glosses in a sign language video. Enhancing the generalization ability of CSLR's visual feature extractor is a worthy area of investigation. In this paper, we model glosses as priors that help to learn more generalizable visual features. Specifically, the signer-invariant gloss feature is extracted by a pre-trained gloss BERT model. Then we design a gloss prior guidance network (GPGN). It contains a novel parallel densely-connected temporal feature extraction (PDC-TFE) module for multi-resolution visual feature extraction. The PDC-TFE captures the complex temporal patterns of the glosses. The pre-trained gloss feature guides the visual feature learning through a cross-modality matching loss. We propose to formulate the cross-modality feature matching into a regularized optimal transport problem, it can be efficiently solved by a variant of the Sinkhorn algorithm. The GPGN parameters are learned by optimizing a weighted sum of the cross-modality matching loss and CTC loss. The experiment results on German and Chinese sign language benchmarks demonstrate that the proposed GPGN achieves competitive performance. The ablation study verifies the effectiveness of several critical components of the GPGN. Furthermore, the proposed pre-trained gloss BERT model and cross-modality matching can be seamlessly integrated into other RGB-cue-based CSLR methods as plug-and-play formulations to enhance the generalization ability of the visual feature extractor.

Extracting multilane roads from OpenStreetMap through graph convolutional neural network and road mesh relationship analysis

Multilane roads in vector data are comprised of parallel line features that represent the same road. The extraction of these features is crucial for updating maps and cartographic generalization. This article addresses previous object-level extraction limitations by introducing a mesh-based method that employs a GCNN model to classify them based on geometric features and proximity relationship. A bidirectional region growing clustering method (BRGC) is designed to cluster the identified multilane road meshes belonging to the same road into groups considering their spatial relationships. The object-level multilane roads are extracted based on the clustering results. Validation using road data from OpenStreetMap in Chengdu and Zhengzhou achieved a classification f1-score >90% for multilane road mesh. Moreover, the object-level multilane detection precision and recall were 88.05% and 92.72% on the Chengdu dataset, and 86.99% and 91.37% on the Zhengzhou dataset, which demonstrates the proposed method effectively extracts multilane roads at the object level, aligning with real-world scenarios and human cognition.

PFFN: A Parallel Feature Fusion Network for Remaining Useful Life Early Prediction of Lithium-ion Battery

PFFN: A Parallel Feature Fusion Network for Remaining Useful Life Early Prediction of Lithium-ion Battery

VRT: A Video Restoration Transformer.

Video restoration aims to restore high-quality frames from low-quality frames. Different from single image restoration, video restoration generally requires to utilize temporal information from multiple adjacent but usually misaligned video frames. Existing deep methods generally tackle with this by exploiting a sliding window strategy or a recurrent architecture, which are restricted by frame-by-frame restoration. In this paper, we propose a Video Restoration Transformer (VRT) with parallel frame prediction ability. More specifically, VRT is composed of multiple scales, each of which consists of two kinds of modules: temporal reciprocal self attention (TRSA) and parallel warping. TRSA divides the video into small clips, on which reciprocal attention is applied for joint motion estimation, feature alignment and feature fusion, while self attention is used for feature extraction. To enable cross-clip interactions, the video sequence is shifted for every other layer. Besides, parallel warping is used to further fuse information from neighboring frames by parallel feature warping. Experimental results on five tasks, including video super-resolution, video deblurring, video denoising, video frame interpolation and space-time video super-resolution, demonstrate that VRT outperforms the state-of-the-art methods by large margins (up to 2.16dB) on fourteen benchmark datasets. The codes are available at https://github.com/JingyunLiang/VRT.

Modeling of Multiple Spatial-Temporal Relations for Robust Visual Object Tracking.

Recently, one-stream trackers have achieved parallel feature extraction and relation modeling through the exploitation of Transformer-based architectures. This design greatly improves the performance of trackers. However, as one-stream trackers often overlook crucial tracking cues beyond the template, they prone to give unsatisfactory results against complex tracking scenarios. To tackle these challenges, we propose a multi-cue single-stream tracker, dubbed MCTrack here, which seamlessly integrates template information, historical trajectory, historical frame, and the search region for synchronized feature extraction and relation modeling. To achieve this, we employ two types of encoders to convert the template, historical frames, search region, and historical trajectory into tokens, which are then collectively fed into a Transformer architecture. To distill temporal and spatial cues, we introduce a novel adaptive update mechanism, which incorporates a thresholding component and a local multi-peak component to filter out less accurate and overly disturbed tracking cues. Empirically, MCTrack achieves leading performance on mainstream benchmark datasets, surpassing the most advanced SeqTrack by 2.0% in terms of the AO metric on GOT-10k. The code is available at https://github.com/wsumel/MCTrack.

Multi-frame network feature fusion model and self-attention mechanism for vehicle lane line detection

The traditional lane detection networks mainly use independent single frame images to extract features first and then detect them, which cannot deal with the scene with complex background well. Therefore, this paper proposes a lane parallel detection network based on multi-frame network feature fusion model and self-attention mechanism according to the scene characteristics that vehicles can obtain continuous images during normal driving. Firstly, a parallel feature extraction structure is designed. On the one hand, a single frame network with high precision is used to extract the features of the current frame. On the other hand, a lightweight multi-frame network is designed to extract features of low-resolution multi-frame temporal images. And the recurrent neural network module is used to fuse the extracted temporal features and obtain multi-frame features. Self-attention mechanism can effectively capture the relevant information of internal features. Then the fusion module of single frame feature, multi-frame feature and self-attention feature is designed. The feature map of lane line is output by up-sampling network. The experimental results show that the network in this paper has significant improvement in both objective detection accuracy and subjective effect compared with other methods.

Enhancing Anomaly Detection in Pedestrian Walkways using Improved Sparrow Search Algorithm with Parallel Features Fusion Model

Anomaly detection in pedestrian walkways is a vital research area, widely employed to enhance the safety of the pedestrians. Because of the widespread usage of the video surveillance systems and the increasing number of captured videos, the conventional manual examination of labeling abnormal events is a laborious process. Therefore, an automatic surveillance system to accurately detect anomalies becomes essential among computer vision researchers. Presently, the development of deep learning (DL) models has gained significant interest in different computer vision processes namely object classification and object detection, and these applications were depending on supervised learning that required labels. This article develops an Improved Meta-heuristic with Parallel Features Fusion Model for Anomaly Detection in Pedestrian Walkways (IMPFF-ADPW) method. The main aim of the IMPFF-ADPW approach is to recognize the existence of anomalies in pedestrian walkways. To obtain this, the IMPFF-ADPW method applies a joint bilateral filter (JBF) for the process of noise removal. Besides, a parallel fusion process comprising NasNet Mobile and Darknet-53 models can be utilized for feature extraction. For the anomaly detection method, the deep autoencoder (DAE) model is applied and its hyperparameters are finetuned by using an improved sparrow search algorithm (ISSA). A wide of experimental outcomes can be applied to the UCSD database to illustrate the betterment of the IMPFF-ADPW methodology. The simulation values indicated the enhanced performance of the IMPFF-ADPW method over other existing techniques.

Object Detection in Remote Sensing Images With Parallel Feature Fusion and Cascade Global Attention Head

Object Detection in Remote Sensing Images With Parallel Feature Fusion and Cascade Global Attention Head

Computer motherboard assembly defect detection using parallel feature extraction and progressive feature fusion

Computer motherboard assembly defect detection using parallel feature extraction and progressive feature fusion

SWFormer: Stochastic Windows Convolutional Transformer for Hybrid Modality Hyperspectral Classification.

Joint classification of hyperspectral images with hybrid modality can significantly enhance interpretation potentials, particularly when elevation information from the LiDAR sensor is integrated for outstanding performance. Recently, the transformer architecture was introduced to the HSI and LiDAR classification task, which has been verified as highly efficient. However, the existing naive transformer architectures suffer from two main drawbacks: 1) Inadequacy extraction for local spatial information and multi-scale information from HSI simultaneously. 2) The matrix calculation in the transformer consumes vast amounts of computing power. In this paper, we propose a novel Stochastic Window Transformer (SWFormer) framework to resolve these issues. First, the effective spatial and spectral feature projection networks are built independently based on hybrid-modal heterogeneous data composition using parallel feature extraction, which is conducive to excavating the perceptual features more representative along different dimensions. Furthermore, to construct local-global nonlinear feature maps more flexibly, we implement multi-scale strip convolution coupled with a transformer strategy. Moreover, in an innovative random window transformer structure, features are randomly masked to achieve sparse window pruning, alleviating the problem of information density redundancy, and reducing the parameters required for intensive attention. Finally, we designed a plug-and-play feature aggregation module that adapts domain offset between modal features adaptively to minimize semantic gaps between them and enhance the representational ability of the fusion feature. Three fiducial datasets demonstrate the effectiveness of the SWFormer in determining classification results.

CAM-QUS guided self-tuning modular CNNs with multi-loss functions for fully automated breast lesion classification in ultrasound images

Objective. Breast cancer is the major cause of cancer death among women worldwide. Deep learning-based computer-aided diagnosis (CAD) systems for classifying lesions in breast ultrasound images can help materialise the early detection of breast cancer and enhance survival chances. Approach. This paper presents a completely automated BUS diagnosis system with modular convolutional neural networks tuned with novel loss functions. The proposed network comprises a dynamic channel input enhancement network, an attention-guided InceptionV3-based feature extraction network, a classification network, and a parallel feature transformation network to map deep features into quantitative ultrasound (QUS) feature space. These networks function together to improve classification accuracy by increasing the separation of benign and malignant class-specific features and enriching them simultaneously. Unlike the categorical crossentropy (CCE) loss-based traditional approaches, our method uses two additional novel losses: class activation mapping (CAM)-based and QUS feature-based losses, to capacitate the overall network learn the extraction of clinically valued lesion shape and texture-related properties focusing primarily the lesion area for explainable AI (XAI). Main results. Experiments on four public, one private, and a combined breast ultrasound dataset are used to validate our strategy. The suggested technique obtains an accuracy of 97.28%, sensitivity of 93.87%, F1-score of 95.42% on dataset 1 (BUSI), and an accuracy of 91.50%, sensitivity of 89.38%, and F1-score of 89.31% on the combined dataset, consisting of 1494 images collected from hospitals in five demographic locations using four ultrasound systems of different manufacturers. These results outperform techniques reported in the literature by a considerable margin. Significance. The proposed CAD system provides diagnosis from the auto-focused lesion area of B-mode BUS images, avoiding the explicit requirement of any segmentation or region of interest extraction, and thus can be a handy tool for making accurate and reliable diagnoses even in unspecialized healthcare centers.