Search

Person Re-identification (Re-ID) technology aims to solve the matching problem of the same pedestrians at different times and places, which has important application value in the field of public safety. At present, most scholars focus on designing complex models to improve the accuracy of Re-ID, but the high complexity of the model further restricts the practical application of Re-ID algorithm. To solve the above problems, this paper designs a lightweight Self-selective Receptive Field (SRF) block instead of directly designing complex models. Specifically, the module can be plug-and-play on the general backbone network, so as to significantly improve the performance of Re-ID while effectively controlling the amount of its own parameter and calculation: (1) the SRF block encodes pedestrian targets and image contexts at different scales by constructing pyramidal convolution group and allows the module to independently select the size of the receptive field through training by means of self-adaptive weighting; (2) in order to reduce the complexity of SRF block, we introduce a "channel scaling factor" and design a "grouped convolution operation" by constraining the channels of the feature map and changing the structure of the convolution kernel respectively. Experiments on multiple datasets show that SRF Network (SRFNet) for Re-ID can achieve a good balance between performance and complexity, which fully demonstrates the effectiveness of SRF block. [ABSTRACT FROM AUTHOR]

Existing person re-identification methods are difficult to generalize to unseen datasets because of the domain gaps. The key to solving this problem is to extract domain-invariant and discriminative features. In this paper, we propose a normalization and enhancement (NE) module that can effectively suppress domain gaps and enhance pedestrian features without any target domain data, thereby enhancing the generalization ability of the model. NE module consists of a residual connection and a channel attention (CA) block. The residual connection is designed to suppress the domain gaps while retaining discriminative information. The proposed CA block embeds spatial information into the channel dimension and models the dependencies between channels to enhance pedestrian features. In addition, the NE loss is designed to constrain the training of NE module to extract domain-invariant features with excellent distribution. Ablation experiments were conducted on Market-1501, DukeMTMC-reID, CUHK03-NP and MSMT17. Experimental results confirmed the superiority of the proposed method, the mAP reached a maximum of 47.3% in cross-domain scenarios. [ABSTRACT FROM AUTHOR]

Aiming at the problem of low accuracy of person re-identification (Re-ID) algorithm caused by occlusion, low distinctiveness of person features and unclear detail features in complex environment, we propose a Re-ID method based on fine-grained feature fusion and self-attention mechanism. First, we design a dilated non-local module (DNLM), which combines dilated convolution with the non-local module and embeds it between layers of the backbone network, enhancing the self-attention and receptive field of the model and improving the performance on occlusion tasks. Second, the fine-grained feature fusion screening module (3FSM) is improved based on the outlook attention module, which can realize adaptive feature selection and enhance the recognition ability to similar samples of the model. Finally, combined with the feature pyramid in the field of object detection, we propose a multi-scale feature fusion pyramid (MFFP) to improve the Re-ID tasks, in which we use different levels of features to perform feature enhancement. Ablation and comprehensive experiment results based on multiple datasets validate the effectiveness of our proposal. The mean Average Precision (mAP) of Market1501 and DukeMTMC-reID is 92.5 and 87.7%, and Rank-1 is 95.1 and 91.1% respectively. Compared with the current mainstream Re-ID algorithm, our method has excellent Re-ID performance. [ABSTRACT FROM AUTHOR]

Multi-object tracking (MOT) aims to locate and identify objects in videos. As deep learning brings excellent performances to object detection, the tracking-by-detection (TBD) has gradually become a mainstream tracking framework. However, some drawbacks still exist in the current TBD framework: (1) inaccurate prediction of the bounding boxes would occur in the detection part, which is caused by overlooking the actual pedestrian ratio in the surveillance scene. (2) The width of the bounding boxes in the next frame might be indirectly predicted by the aspect ratio, which increases the error of width prediction in the motion prediction part. (3) Association is only performed for high-confidence detection boxes, and the low-confidence boxes caused by occlusion are discarded in the data association part, resulting in fragmentation of trajectories. To address the above issues, we propose a multi-target tracking model incorporating motion estimation and multi-stage association (MEMA). First, the aspect ratio of the ground-true bounding box is introduced to improve the fit of the detection and the ground-true bounding box, and we design the elliptical Gaussian kernel to improve the positioning accuracy of the object center point. Then, the prediction state vector of the Kalman filter is modified to predict the width and its corresponding velocity directly. It can reduce the width error of the prediction box and eliminate the velocity error of the motion estimation, which leads to a more pedestrian-friendly prediction bounding box. Finally, we propose a multi-stage association strategy to correlate different confidence boxes. Without using the appearance feature, the strategy can reduce the impact of occlusion and improve the tracking performance. On the MOT17 test set, the method proposed in this paper achieves a MOTA of 74.3% and an IDF1 of 72.4%, outperforming the current SOTA. [ABSTRACT FROM AUTHOR]

Understanding user behavior via IP addresses is a crucial measure towards numerous pragmatic IP-based applications, including online content delivery, fraud prevention, marketing intelligence, and others. While profiling IP addresses through methods like IP geolocation and anomaly detection has been thoroughly studied, the function of an IP address—e.g., whether it pertains to a private enterprise network or a home broadband—remains underexplored. In this work, we initiate the first attempt to address the IP usage scenario classification problem. We collect data consisting of IP addresses from four large-scale regions. A novel continuous neural tree-based ensemble model is proposed to learn IP assignment rules and complex feature interactions. We conduct extensive experiments to evaluate our model in terms of classification accuracy and generalizability. Our results demonstrate that the proposed model is capable of efficiently uncovering significant higher-order feature interactions that enhance IP usage scenario classification, while also possessing the ability to generalize from the source region to the target one. [ABSTRACT FROM AUTHOR]

Deep neural networks (DNNs) are vulnerable to adversarial examples that fool the models with tiny perturbations. Although adversarial attacks have achieved incredible attack success rates in the white-box setting, most existing adversaries often exhibit weak transferability in the black-box setting, especially for models with defense mechanisms. In this work, we reveal the cross-model channel redundancy and channel invariance of DNNs and thus propose two channel-augmented methods to improve the transferability of adversarial examples, namely, the channel transformation (CT) method and the channel-invariant Patch (CIP) method. Specifically, channel transformation shuffles and rewrites channels to enhance cross-model feature redundancy in convolution, and channel-invariant patches distinctly weaken different channels to achieve loss-preserving transformation. We compute the aggregated gradients of the transformed dataset to create adversaries with higher transferability. In addition, the two proposed methods can be naturally combined with each other and with almost all other gradient-based methods to further improve performance. Empirical results on the ImageNet dataset demonstrate that our attack methods exhibit higher transferability and achieve higher attack success rates than state-of-the-art gradient-based attacks. Specifically, our attack improves the average attack success rate from 86.9% to 91.0% on normally trained models and from 44.6% to 68.3% on adversarially trained models. [ABSTRACT FROM AUTHOR]

Multi-object tracking (MOT) is mainly used for detecting and tracking the object on multi-cameras, which is widely applied in intelligent video surveillance and intelligent security. The process of MOT generally involves three import parts: feature extracting, multi-task learning and object matching. Unfortunately, the existed methods still have some drawbacks. Firstly, the feature extracting module cannot effectively fuse the shallow and deep features. What's more, the multi-task learning module cannot strike a good balance between the detection and re-identification. In addition, the object matching module associates with pedestrian by using a traditional method rather than training a model. For these problems, we propose a method of joint detection and association (JDA) for end-to-end multi-object tracking network, which involves the multi-scale feature extraction and the learnable object association. It first combines a feature extraction backbone based on multi-scale feature fusion and a point-based multi-task object detection branch, to solve the task of feature extraction and object detection. Then, a learnable object motion association module is embedded, which uses the historical frames information to infer the position of the object, and associate the object identity between previous frames and subsequent frames. In addition, the JDA can be end-to-end trained when handling the detection and matching tasks. The proposed JDA is evaluated through a series of experiments on MOT16 and MOT17. The results shows that JDA the existing methods in terms of precision and stability of MOT. [ABSTRACT FROM AUTHOR]

To enhance the fatigue performance of the new TM210A steel and analyze the influence of shot peening on such properties. The surface test was used to analyze the effects of shot peening on surface morphology, roughness and residual stress, and rotational flexural fatigue tests were carried out to observe fatigue fracture morphology and analyze fatigue performance of the base material and shot-peened specimen. After shot peening, the surface of the specimen had obvious plastic deformation, the surface roughness of the specimen increased obviously and Ra value increased from 0.47 to 1.33 μm, and the residual compressive stress layer of about 210 μm was formed. The influence mechanism of the residual stress, generated by shot peening, on the fatigue performance of the new TM210A steel was analyzed based on the finite element calculation. The results reveal that the fatigue life and fatigue limit of the new TM210A steel could be significantly enhanced by shot peening, the fatigue limit of the base material specimen was 617 MPa, the fatigue limit of the shot-peened specimen was 847 MPa, and the increase was 37%. The influence mechanism of shot peening on the fatigue performance of the new TM210A steel was that the residual stress generated by shot peening moved the fatigue source from the specimen surface to the interior. It could also be identified that the smaller the external load stress, the deeper the fatigue source is from the specimen surface, the smaller the possible range, and the more obvious the enhancement of fatigue life. [ABSTRACT FROM AUTHOR]

There are notable style differences among person re-identification (ReID) datasets, such as brightness, tone, resolution, background, and clothing style, that result in serious challenges for cross-domain person ReID. Two methods are usually used to solve these problems. One is to remove style differences between datasets by applying specific modules, such as instance normalization (IN). However, this method will filter out large amounts of valuable information for ReID. The other is to use person attributes as auxiliary information, but this method does not deeply explore the relationship between attribute features and global features, resulting in underutilized attribute information. We propose the domain-invariant feature extraction and fusion (DFEF), which consists of the attention and style normalization (ASN) and the attribute feature extraction and fusion (AFEF). The ASN module integrates spatial and channel attention modules on the basis of the IN layer to effectively remove the style differences between datasets and recovers the filtered-out information, which is useful for ReID. The AFEF module includes the attribute branch and the feature fusion module. For the attribute branch, we embed the convolutional block attention module (CBAM) into the attribute branch and adopt the multi-label focal loss (MLFL) to improve the accuracy of attribute recognition. For the feature fusion module, we propose the dispersion reweighting strategy to explore the correlation between attribute features and global features. The proposed DFEF method achieves 30.1% and 35.0% mAP on Market-1501 → DukeMTMC-reID and DukeMTMC-reID → Market-1501, respectively. [ABSTRACT FROM AUTHOR]

The successful implementation of bioelectronic devices attached to living organism hinges on a number of material and device characteristics, including not only electrical and mechanical performances to gather physiological signals from living organism thus enabling status monitoring, but also permeability or breathability for gas/nutrient exchange between living organisms and surroundings to ensure minimum perturbation of the intrinsic biological function. However, most bioelectronic devices built on planar polymeric substrates, such as polydimethylsiloxane (PDMS), polyurethane (PU), and polyimide (PI), lack efficient gas permeability, which may hinder the emission of volatile compounds from the surface of living organism, affecting the natural metabolism and reducing the comfort of wearing. Thus, achieving permeability or breathability in bioelectronic devices is a significant challenge. Currently, the devices made of gas-permeable materials with porous structures, that combine electronic components with daily garments, such as fibric and textile, offer exciting opportunities for breathable electronics. In this review, several types of gas-permeable materials with their synthesis and processing routes are outlines. Then, two methods for measuring water vapor transmission rate of materials are discussed in depth. Finally, recent progress in the use of gas-permeable materials for the applications of plant- and skin-attached electronics is summarized systematically. [ABSTRACT FROM AUTHOR]

Person searches generally involve three important parts: person detection, feature extraction and identity comparison. However, a person search integrating detection, extraction and comparison has the two following drawbacks. First, the accuracy of detection will affect the accuracy of comparison. Second, it is difficult to achieve real-time results in real-world applications. To solve these problems, we propose a multitask joint framework for real-time person search (MJF) that optimizes person detection, feature extraction and identity comparison. For the person detection module, we propose the YOLOv5-GS model, which is trained with a person dataset. YOLOv5-GS combines the advantages of the Ghostnet and the squeeze-and-excitation block and improves the speed of person detection. For the feature extraction module, we design a model adaptation architecture, which can select different networks according to the number of people. It can balance the relationship between accuracy and speed. For identity comparison, we propose a 3D pooled table and a matching strategy to improve identification accuracy. On the condition of 1920 × 1080-resolution video and a 200-ID table, the IR and the FPS achieved by our method reach 82.69% and 25.14, respectively. Therefore, the MJF can achieve real-time person search. [ABSTRACT FROM AUTHOR]

In the monitoring scene, parameters of different cameras are vary greatly, which makes Person re-identification (Re-ID) tasks extremely susceptible to factors such as scale, blur, and occlusion. To alleviate the these problems, this paper proposes a Dense Feature Pyramid Network (DFPN), which can converge to a better performance without pretraining. To be more specific, DFPN is composed of three main parts. First, a new Residual Convolutional Block (RCB) is designed by referring to the construction method of ResBlock. Taking RCB as a basic unit and combining it with the convolution layer structure of VGGNet, we construct the backbone RVNet (Residual VGGNet) to realize the rapid convergence of the network and solve the disappearance of the gradient. Second, based on Feature Pyramid Network, we design the Dense Pyramid Fusion Module by integrating the connection mode of DenseNet, which aims at the improvement of the richness and scale diversity of feature maps by taking semantic information and detail information into account. Finally, to increase the receptive field of the feature map, we introduce an improved retinal receptive field structure Improved RFB (IRFB) on the basis of Receptive Field Block (RFB), which can effectively solve the problem of pedestrian occlusion. In experiments on the public datasets Market1501, DukeMTMC-reID and Occluded-Duke, the Rank-1 accuracy can reach 94.12%, 87.25% and 51.72% with pretraining, respectively. A series of ablation experiments and comparative experiments have proved the effectiveness of our modules and overall scheme. [ABSTRACT FROM AUTHOR]

This paper addresses a two-agent scheduling problem with due date assignment under multitasking environment, in which the due dates of the jobs from the first agent are decision variables to be determined using the unrestricted (usually referred to as DIF) due date assignment method. Each agent requests the processing of its own set of jobs on a machine and wishes to minimize a certain scheduling criterion related to the completion times of its jobs only. Under multitasking, when a job (primary job) is processed, it is inevitably interrupted by other jobs (waiting jobs) that are available but unfinished, and the amount of time that each waiting job interrupting the primary job is a linear function of the remaining processing time of the waiting job. The overall objective is to determine the optimal primary job sequence along with the due dates of the jobs from the first agent as to minimize the weighted sum of the due date assignment cost and weighted number of late jobs from the first agent, while maintaining the total completion time of the jobs from the second agent not exceeding a given threshold. We show that the problem is NP -hard, devise a pseudo-polynomial time dynamic programming algorithm, establishing that it is NP -hard in the ordinary sense, and demonstrate that it admits a fully polynomial-time approximation scheme. [ABSTRACT FROM AUTHOR]

Septuple-atomic-layer MoSi2N4 has emerged as a promising and non-precious energy conversion material due to its stable and unique crystal structures. Unfortunately, the MoSi2N4 monolayer shows an inert catalytic active feature due to the large positive Gibbs free energy (ΔGH). To overcome the main drawback, an effective alternative is to design MoSi2N4 with different modifications. The results show that the electronic properties can be significantly tailored with the atomic vacancies, substitutions and biaxial strains with the band gap reduced to near zero eV when B and O dopants are incorporated into MoSi2N4 monolayer. Moreover, the continuous decrease of band gap can also be obtained by applying the tensile strains on MoSi2N4 monolayer. The ELF and charge transfer analyses show that the modifications induce various cation–anion bond features for the MoSi2N4. The changes of bond features greatly enhance the hydrogen adsorption on the MoSi2N4, rendering a suitable ΔGH (near 0 eV) for HER. The obvious enhancement is owing to the H-1 s orbital splitting into two parts located below and above Fermi energy due to the strong hybridization between the hydrogen atom and catalyst. These results suggest the great potential of such modified MoSi2N4 monolayer as highly efficient HER electrocatalyst for practical applications. [ABSTRACT FROM AUTHOR]

Highlights: The lateral-line-like underwater mechanical sensor (LUMS) realizes the imitation of the structure and function of the lateral line. The detection range of water depth can be controlled by adjusting the size of the graphene/Ecoflex Janus film on LUMS. The maximum measured depth is 1.8 m. Similar to the fish, the mechanical stimuli from land and water can be sensitively captured by LUMS in real time.Underwater exploration has been an attractive topic for understanding the very nature of the lakes and even deep oceans. In recent years, extensive efforts have been devoted to developing functional materials and their integrated devices for underwater information capturing. However, there still remains a great challenge for water depth detection and vibration monitoring in a high-efficient, controllable, and scalable way. Inspired by the lateral line of fish that can sensitively sense the water depth and environmental stimuli, an ultrathin, elastic, and adaptive underwater sensor based on Ecoflex matrix with embedded assembled graphene sheets is fabricated. The graphene structured thin film is endowed with favourable adaptive and morphable features, which can conformally adhere to the structural surface and transform to a bulged state driven by water pressure. Owing to the introduction of the graphene-based layer, the integrated sensing system can actively detect the water depth with a wide range of 0.3–1.8 m. Furthermore, similar to the fish, the mechanical stimuli from land (e.g. knocking, stomping) and water (e.g. wind blowing, raining, fishing) can also be sensitively captured in real time. This graphene structured thin-film system is expected to demonstrate significant potentials in underwater monitoring, communication, and risk avoidance. [ABSTRACT FROM AUTHOR]

This study addresses the multitasking scheduling problems with batch distribution and due date assignment (DDA). Compared with classical scheduling problems with due date-related optimization functions, the job due dates are decision variables rather than given parameters. The jobs completed are distributed in batches, and the sizes of all batches are identical, which may be bounded or unbounded. The jobs in every batch are scheduled one by one. Each batch incurs a fixed cost. Under multitasking environment, it allows the machine to put an uncompleted job on hold and turn to another uncompleted job. The goal is to identify the optimal primary job sequence, the optimal job due dates, and the optimal batch production and distribution strategy so that one of the following two optimization functions is minimised: the total cost composed of the earliness penalty, DDA cost, tardiness penalty and batch distribution cost, and the total cost composed of the earliness penalty, weighted number of late jobs, DDA cost and batch distribution cost. We devise efficient exact algorithms for the problems we consider, and perform numerical experiments to check how multitasking affects the scheduling cost or value, the results of which can assist decision-makers to justify the extent to put to use or refrain from multitasking. [ABSTRACT FROM AUTHOR]