Person re-identification (re-id) plays a vital role in surveillance and forensics application. Since the labeled images for person re-id task is limited, the generalization ability of existed person re-id models is poor. On the other hand, images of different classes (pedestrian and non-pedestrian images) share some general features. To this end, this paper aims to improve the performance of person re-id by designing a relearning network which can learn domain-specific features and general features simultaneously. The proposed relearning network consists of a pretrained backbone network which provides the general features, and several attention-based subnetworks that learn domain-specific features from general features of different levels. Besides, we propose a coarse-fine loss to improve the generalization of person re-id model by making full use of the massive labeled non-pedestrian images. Experimental results on the publicly available Market-1501, DukeMTMC-reID and CUHK03 pedestrian re-id datasets demonstrate the effectiveness of the proposed relearning network and coarse-fine loss.INDEX TERMS Attention, coarse-fine, domain-specific features, general features.
SummaryThe rapid update of computing power leads to exponential data traffic growth, and the incidence of network attacks is also increasing. It is significantly important to analyze and predict network traffic accurately in the early stage and take corresponding preventive measures. The existing network flow integrated forecasting models still have some bottlenecks that are difficult to solve, for example, the slow optimization speed of modal decomposition parameters, easy falling into local optimal solutions, the slow convergence speed of the training process, and poor generalization capability. In this paper, particle swarm optimization (PSO) is utilized to improve the parameters selection process of the variational mode decomposition (VMD) algorithm and the extreme learning machine (ELM) algorithm. First, the PSO‐VMD combined with multi‐scale permutation entropy (MPE) is utilized to decompose the original network flow, and multiple eigenmode components are obtained. Second, the PSO‐ELM is utilized to train the network traffic prediction model, and the PSO parameters in PSO‐ELM are updated through adaptive weight adjustment and synchronous learning factors to increase the training and prediction speed, and the component prediction results are reconstructed to get a high‐precision network flow forecasting result. Finally, through the prediction and verification of the public network flow data of the WIDE backbone, the result of this experiment indicates that the VMD‐PSO‐ELM can break through the bottlenecks of slow optimization speed of VMD decomposition parameters, reduce the computational complexity of ELM, accelerate the convergence speed, and increase the forecasting accuracy.
Foreground detection is an essential step in computer vision and video processing. Accurate foreground object extraction is crucial for subsequent high-level tasks such as target recognition and tracking. Although many foreground detection algorithms have been proposed, foreground detection in complex scenes is still a challenging problem. This paper presents a foreground detection algorithm based on superpixel and semantic segmentation. It first uses multiscale superpixel segmentation to obtain the initial foreground mask. At the same time, a semantic segmentation network is applied to separate potential foreground objects, and then use the defined rules to combine the results of superpixel and semantic segmentation to get the final foreground object. Finally, the background model is updated with the refined foreground result. Experiments on the CDNet2014 dataset demonstrate the effectiveness of the proposed algorithm, which can accurately segment foreground objects in complex scenes.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.