Person search aims to locate target individuals in large image databases captured by multiple non-overlapping cameras. Existing models primarily rely on spatial feature extraction to capture fine-grained local details...
Person search aims to locate target individuals in large image databases captured by multiple non-overlapping cameras. Existing models primarily rely on spatial feature extraction to capture fine-grained local details, which is vulnerable to background clutter and occlusions and leads to unstable feature representations. To address the issues, we propose a Dynamic Frequency Selection and Spatial Interaction Fusion Network (PS-DFSI), marking the first attempt to introduce frequency decoupling and selection into person search. By integrating frequency and spatial features, PS-DFSI enhances feature expressiveness and robustness. Specifically, it comprises two core modules: the Dynamic Frequency Selection Module (DFSM) and the Spatial Frequency Interaction Module (SFIM). DFSM decouples feature maps into low-frequency and high-frequency components using learnable low-pass and high-pass filters, and a frequency selection modulator emphasizes key frequency components via channel attention. SFIM refines local details by fusing frequency-enhanced features with high-level semantic representations, leveraging multi-scale receptive fields and cross-feature attention for efficient spatial-frequency integration. Extensive experiments on CUHK-SYSU and PRW demonstrate that PS-DFSI significantly improves person search performance, validating its effectiveness and robustness.
Modern datacenter servers hosting popular Internet services face significant and multi-facet challenges in performance and power control. The user-perceived performance is the result of a complex interaction of comple...
详细信息
Modern datacenter servers hosting popular Internet services face significant and multi-facet challenges in performance and power control. The user-perceived performance is the result of a complex interaction of complex workloads in a very complex underlying system. Highly dynamic and bursty workloads of Internet services fluctuate over multiple time scales, which has a significant impact on processing and power demands of datacenter servers. High-density servers apply virtualization technology for capacity planning and system manageability. Such virtuMized computer systems are increasingly large and complex. This paper surveys representative approaches to autonomic performance and power control on virtualized servers, which control the quality of service provided by virtualized resources, improve the energy efficiency of the underlying system, and reduce the burden of complex system management from human operators. It then presents three designed self-adaptive resource management techniques based on machine learning and control for percentile-based response time assurance, non-intrusive energy-efficient performance isolation, and joint performance and power guarantee on virtualized servers. The techniques were implemented and evaluated in a testbed of virtualized servers hosting benchmark applications. Finally, two research trends are identified and discussed for sustainable cloud computing in green datacenters.
User behavioral analysis is expected to act as a promising technique for identity theft detection in the Internet. The performance of this paradigm extremely depends on a good individual-level user behavioral model. S...
详细信息
Visible‐infrared person re‐identification(VI‐ReID)is a supplementary task of single‐modality re‐identification,which makes up for the defect of conventional re‐identification under insufficient *** is more chall...
详细信息
Visible‐infrared person re‐identification(VI‐ReID)is a supplementary task of single‐modality re‐identification,which makes up for the defect of conventional re‐identification under insufficient *** is more challenging than single‐modality ReID because,in addition to difficulties in pedestrian posture,camera shoot-ing angle and background change,there are also difficulties in the cross‐modality *** works only involve coarse‐grained global features in the re‐ranking calculation,which cannot effectively use fine‐grained ***,fine‐grained features are particularly important due to the lack of information in cross‐modality re‐*** this end,the Q‐center Multi‐granularity K‐reciprocal Re‐ranking Algorithm(termed QCMR)is proposed,including a Q‐nearest neighbour centre encoder(termed QNC)and a Multi‐granularity K‐reciprocal Encoder(termed MGK)for a more comprehensive feature *** converts the probe‐corresponding modality features into gallery corresponding modality features through modality transfer to narrow the modality *** takes a coarse‐grained mutual nearest neighbour as the dominant and combines a fine‐grained nearest neighbour as a supplement for similarity *** experiments on two widely used VI‐ReID benchmarks,SYSU‐MM01 and RegDB have shown that our method achieves state‐of‐the‐art ***,the mAP of SYSU‐MM01 is increased by 5.9%in all‐search mode.
By constructing a list of IF-THEN rules, the traditional ant colony optimization(ACO) has been successfully applied on data classification with not only a promising accuracy but also a user comprehensibility. However,...
详细信息
ISBN:
(纸本)9781538619797;9781538619780
By constructing a list of IF-THEN rules, the traditional ant colony optimization(ACO) has been successfully applied on data classification with not only a promising accuracy but also a user comprehensibility. However, as the collected data to be classified usually contain large volumes and redundant features, it is challenging to further improve the classification accuracy and meanwhile reduce the computational time for *** paper proposes a novel hybrid mutual information based ant colony algorithm(mrAM+) for classification. First, a maximum relevance minimum redundancy feature selection method is used to select the most informative and discriminative attributes in a dataset. Then, we use the enhanced ACO classifier(i.e., AM+)to perform the classification. Experimental results show that the proposed mrAM+ outperforms other seven state-of-art related classification algorithms in terms of accuracy and the size of model.
In this paper, we present the parallel implementation of the traffic microsimulation PMTS (Parallel Microscopic Traffic Simulation) focusing on the performance issues. The parallelization of PMTS is domain decompositi...
详细信息
ISBN:
(纸本)9781427629807
In this paper, we present the parallel implementation of the traffic microsimulation PMTS (Parallel Microscopic Traffic Simulation) focusing on the performance issues. The parallelization of PMTS is domain decomposition, which means that each processor of the PC cluster is responsible for a different geographical area of the simulation region. We describe the transportation network graph partition and information exchange between domains. We demonstrate the time cost mathematics models for PMTS: the vehicle generation, vehicle position calculation, and vehicle information exchange between domains. The workload balance is obtained by adjusting the boundary lines according to the relative load of adjacent subnetworks. All these works have been proved to be effective when PMTS put into use and the experiment results are also provided which match our analysis.
Effectiveness is the most important factor considered in the ranking models yielded by algorithms of learning to rank (LTR). Most of the related ranking models only focus on improving the average effectiveness but ign...
详细信息
The upper bounds on lifetime of three dimensional extended Time hopping impulse radio Ultrawide band (TH-IR UWB) sensor networks are derived using percolation theory arguments. The TH-IR UWB sensor network consists of...
详细信息
The upper bounds on lifetime of three dimensional extended Time hopping impulse radio Ultrawide band (TH-IR UWB) sensor networks are derived using percolation theory arguments. The TH-IR UWB sensor network consists of n sensor nodes distributed in a cube of edge length n1/3 according to a Poisson point process of unit intensity. It is shown that for such a static three dimensional extended TH-IR UWB sensor network, the upper bound on the lifetime is of order O(n-1), while in the ideal case, the upper bound on the lifetime is longer than that of a static network by a factor of n 2/3. Therefore sensor nodes moving randomly in the deployment area can improve the upper bound on network lifetime. The results also reveal that the upper bounds on network lifetime decrease with the number of nodes n, thus extended THIR UWB sensor networks aren't prone to be employed in large-scale network.
Predictive Business Process Monitoring(PBPM)is a significant research area in Business Process Management(BPM)aimed at accurately forecasting future behavioral *** present,deep learning methods are widely cited in PBP...
详细信息
Predictive Business Process Monitoring(PBPM)is a significant research area in Business Process Management(BPM)aimed at accurately forecasting future behavioral *** present,deep learning methods are widely cited in PBPM research,but no method has been effective in fusing data information into the control flow for multi-perspective process ***,this paper proposes a process prediction method based on the hierarchical BERT and multi-perspective data ***,the first layer BERT network learns the correlations between different category attribute ***,the attribute data is integrated into a weighted event-level feature vector and input into the second layer BERT network to learn the impact and priority relationship of each event on future predicted ***,the multi-head attention mechanism within the framework is visualized for analysis,helping to understand the decision-making logic of the framework and providing visual ***,experimental results show that the predictive accuracy of the framework surpasses the current state-of-the-art research methods and significantly enhances the predictive performance of BPM.
Current semi-supervised learning-based sample selection methods for noisy label image classification typically utilize all clean and noisy samples for model training. However, not all noisy samples contribute positive...
详细信息
暂无评论