Discriminative correlation filters (DCF) have significantly advanced visual target tracking. However, most DCF-based trackers suffer from various challenges such as occlusion, rotation, and background clutters. Theref...
详细信息
Discriminative correlation filters (DCF) have significantly advanced visual target tracking. However, most DCF-based trackers suffer from various challenges such as occlusion, rotation, and background clutters. Therefore, we propose a novel visual tracking framework, which introduces a motion-aware strategy and automatic temporal regularization mechanism into the spatial-temporal regularization correlation filter (STRCF) to improve tracking stability. Specifically, the motion-aware strategy based on the optimal Kalman filter (KF) is used to estimate the possible state of the target for overcoming the instability problem in complex environments. Furthermore, a novel automatic temporal regularization mechanism is proposed to solve the problem of target drift due to overhigh temporal penalty. Compared with STRCF, our method obtains AUC gains of 5.86%, 2.60%, 3.82%, 4.95%, 3.55%, and 1.90% for the occlusion, motion blur, in-plane Rotation, out-of-plane rotation, background clutters, and scale variation attributes on the OTB-2015 datasets, respectively. Extensive experiment results on OTB-2013, DTB-70, and UAV-123 datasets have proven the effectiveness and stability of our method.
Shared bicycles are an environmentally friendly and convenient means of transportation that can be found everywhere in life. However, problems such as over delivery, disorderly occupation and O&M mismatch may occu...
详细信息
Adversarial noise attacks present a significant threat to quantum machine learning (QML) models, similar to their classical counterparts. This is especially true in the current Noisy Intermediate-Scale Quantum era, wh...
详细信息
The Internet of Things (IoT) is increasingly vulnerable to security risks due to new network attacks. Deep learning-based intrusion detection systems (DL-IDS) have emerged as a key solution, but they face challenges l...
详细信息
Dynamic recommendation, focusing on modeling user preference from historical interactions and providing recommendations on current time, plays a key role in many personalized services. Recent works show that pre-train...
详细信息
In this paper, we study the obstacle avoidance problem of second-order nonlinear multi-agent systems (MASs) with directed graph based on event-triggered control. Firstly, the consensus requirement is accomplished by u...
详细信息
Multivariate time series anomaly detection, crucial for ensuring the safety of real-world systems, primarily focuses on extracting characteristics from time series under normal condition, and identifying potential ano...
Multivariate time series anomaly detection, crucial for ensuring the safety of real-world systems, primarily focuses on extracting characteristics from time series under normal condition, and identifying potential anomalies throughout the evaluation process. Recent studies have achieved fruitful progress through mining the spatio-temporal relationships from multivariate time series, however, these approaches mostly neglect the latency among series which could lead to higher false alarm. Granger causality presents a promising solution to extract these inherent time-lagged relationships. Nonetheless, the intricate and dynamic relationships among numerous time series in real-world systems surpass the ability of linear Granger causality. To address this, we extend the linear Granger causality and propose the Granger Causal Former (GCFormer), a novel approach that leverages attention mechanisms to learn the inherent causal spatio-temporal relationships between historical and current timestamps across multiple time series. Specifically, GCFormer develops a Spatio-Mask (SM) to select the top-k most relevant series and a Temporal-Mask (TM) to concentrate attention on more recent historical timestamps. Moreover, to mitigate overfitting and ensure a smooth training process, GCFormer introduces an adjust top-k method and a TM penalty term. We evaluated GCFormer on four real-world benchmark datasets, demonstrating its superior performance over state-of-the-art approaches. Further analysis and a case study highlight the model’s novelty and interpretability.
Textured Nb_(4)AlC_(3)ceramics were rapidly and efficiently prepared by hot forging through spark plasma sintering(SPS).The longitudinal compression ratio of textured Nb_(4)AlC_(3)ceramics was−78.3%,and the lateral ex...
详细信息
Textured Nb_(4)AlC_(3)ceramics were rapidly and efficiently prepared by hot forging through spark plasma sintering(SPS).The longitudinal compression ratio of textured Nb_(4)AlC_(3)ceramics was−78.3%,and the lateral expansion ratio was 32.1%.The grains grew preferentially along the direction perpendicular to the c-axis,forming the texture *** Lotgering orientation factor f(00l)was calculated to be *** thermal conductivity of textured Nb_(4)AlC_(3)ceramics along the c-axis direction(11.23 W·m^(−1)·K^(−1))(25℃)was lower than that of untextured ceramics(13.75 W·m^(−1)·K^(−1))(25℃).The electrical conductivity perpendicular to the c-axis direction reached 4.37×10^(6) S·m^(−1)at room *** ordered layered grains increased the resistance of crack propagation,resulting in a higher fracture toughness parallel to the c-axis direction(9.41 MPa·m^(1/2)),which was higher than that of untextured ceramics(6.88 MPa·m1/2).The Vickers hardness tested at 10 N on the texture top surface(7.18 GPa)was higher than that on the texture side surface(6.45 GPa).
Sentiment analysis mines social media networks like Twitter for views, attitudes, and sentiments. It is presently a popular study subject. The traditional sentiment analysis approach emphasises textual data the most. ...
详细信息
Federated learning (FL) is a promising approach for participants' collaborative learning tasks with cross-silo data. Participants benefit from FL since heterogeneous data can contribute to the generalization of th...
详细信息
Federated learning (FL) is a promising approach for participants' collaborative learning tasks with cross-silo data. Participants benefit from FL since heterogeneous data can contribute to the generalization of the global model while keeping private data locally. However, practical issues of FL, such as security and fairness, keep emerging, impeding its further development. One of the most threatening security issues is the poisoning attack, corrupting the global model by an adversary's will. Recent studies have demonstrated that elaborate model poisoning attacks can breach the existing Byzantine-robust FL solutions. Although various defenses have been proposed to mitigate poisoning attacks, participants will sacrifice learning performance and fairness due to strict regulations. Considering that the importance of fairness is no less than security, it is crucial to explore alternative solutions that can secure FL while ensuring both robustness and fairness. This paper introduces a robust and fair model aggregation solution, Romoa-AFL, for cross-silo FL in an agnostic data setting. Unlike a previous study named Romoa and other similarity-based solutions, Romoa-AFL ensures robustness against poisoning attacks and learning fairness in agnostic FL, which has no assumptions of participants' data distributions and the server's auxiliary dataset.
暂无评论