A5047488004- Video Surveillance and Tracking Methods
- Advanced Image and Video Retrieval Techniques
- Visual Attention and Saliency Detection
- Anomaly Detection Techniques and Applications
- Human Pose and Action Recognition
- Multimodal Machine Learning Applications
- Crystallization and Solubility Studies
- X-ray Diffraction in Crystallography
- Natural product bioactivities and synthesis
- Medicinal Plants and Neuroprotection
- Advanced Neural Network Applications
- Infrared Target Detection Methodologies
- Synthesis and biological activity
- Quinazolinone synthesis and applications
- Image Enhancement Techniques
- Nanoparticle-Based Drug Delivery
- Network Security and Intrusion Detection
- Synthesis and Characterization of Heterocyclic Compounds
- Synthesis and Biological Evaluation
- Neurological Disease Mechanisms and Treatments
- Robotics and Sensor-Based Localization
- Data-Driven Disease Surveillance
- Synthesis of Tetrazole Derivatives
- Face and Expression Recognition
- Genomics, phytochemicals, and oxidative stress
Zhejiang University of Technology
2015-2024
Shaanxi University of Chinese Medicine
2018-2024
Zhejiang Lab
2022
Shaanxi Institute of International Trade & Commerce
2019
Shandong Normal University
2018
Nanjing University of Science and Technology
2013-2014
University of Technology Sydney
2014
Shaanxi Research Association for Women and Family
2006
By decomposing the visual tracking task into two subproblems as classification for pixel category and regression object bounding box at this pixel, we propose a novel fully convolutional Siamese network to solve end-to-end in per-pixel manner. The proposed framework SiamCAR consists of simple subnetworks: one subnetwork feature extraction classification-regression prediction. Different from state-of-the-art trackers like Siamese-RPN, SiamRPN++ SPM, which are based on region proposal, is both...
Siamese network based trackers formulate the visual tracking task as a similarity matching problem. Almost all popular realize learning via convolutional feature cross-correlation between target branch and search branch. However, since size of region needs to be pre-fixed, these base methods suffer from either reserving much adverse background information or missing great deal foreground information. Moreover, global be-tween also largely neglects structure part-level information.In this...
Point cloud completion referring to completing 3D shapes from partial point clouds is a fundamental problem for analysis tasks. Benefiting the development of deep neural networks, researches on have made great progress in recent years. However, explicit local region partition like kNNs involved existing methods makes them sensitive density distribution clouds. Moreover, it serves limited receptive fields that prevent capturing features long-range context information. To solve problems, we...
Abstract Visual tracking of generic objects is one the fundamental but challenging problems in computer vision. Here, we propose a novel fully convolutional Siamese network to solve visual by directly predicting target bounding box an end-to-end manner. We first reformulate task as two subproblems: classification problem for pixel category prediction and regression object status estimation at this pixel. With decomposition, design simple yet effective architecture based framework, termed...
For a given data set, exploring their multi-view instances under clustering framework is practical way to boost the performance. This because that each view might reflect partial information for existing data. Furthermore, due noise and other impact factors, these from different views will enhance mining of real structure feature within set. In this paper, we propose multiple kernel spectral algorithm through on By combining matrix learning optimization into one process framework, can...
Aim: Aralia taibaiensis is a natural medicinal and food plant that rich in triterpenoid saponins with hypoglycaemic, antioxidant, hepatoprotective, anti-gastric ulcer anti-inflammatory effects. This study has significance terms of the anti-aging organ protective effects total (TSAT) D-galactose-induced aging rats. Methods: The saponin composition TSAT was determined quantified by high performance liquid chromatography (HPLC). We consolidated antioxidant enzyme inhibitory activities vitro...
Aralia taibaiensis, a medicinal food plant, and total saponins from its root bark extract inhibit α-glucosidase activity, which is associated with type 2 diabetes; however, the inhibitory mechanism unknown. Furthermore, green extraction technique superior to conventional hot reflux (HRE) needed for rapid easy of A. taibaiensis (TSAT) exploit utilize this resource. Our aim was develop method obtaining TSAT investigate by inhibits α-glucosidase.In study, ultrasound-assisted (UAE) process...
Lingzhu Pulvis is a classic formulation for treating febrile convulsions in children. However,
Cerebral ischaemia-reperfusion injury (CIRI) could worsen the inflammatory response and oxidative stress in brain tissue. According to previous studies, ferulic acid methyl ester (FAME), as extract with strongest comprehensive activity traditional Chinese medicine Huang Hua oil dot herb, has significant anti-oxidative neuroprotective functions, can effectively alleviate CIRI, but its mechanism of action is still unclear.
Siamese-network-based trackers convert the general object tracking as a similarity matching task between template and search region. Using convolutional feature cross correlation (Xcorr) for matching, large number of Siamese are proposed achieved great success. However, due to predefined size target feature, these suffer from either retaining much background information or losing important foreground information. Moreover, global region also largely neglects part-level structural contextual...