Zhiqiang Zhao
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Zhiqiang Zhao has a total of 43 co-authors. Here are the top co-authors by number of shared publications:
- Bin Wu
School of Computer and Big Data Science, Jiujiang University, No. 551, Qianjin East Road, Jiujiang, Jiangxi 332005, China
5 shared publications - Zhuolin Mei
Jiujiang University
4 shared publications Bin Wu
The School of Information Science and Technology Jiujiang University Jiujiang China
3 shared publications- Zongmin Cui
School of Computer and Big Data Science, Jiujiang University, Jiujiang, Jiangxi, China
3 shared publications - Bin Wu
Jiujiang University
2 shared publications - Caicai Zhang
School of Modern Information Technology, Zhejiang Institute of Mechanical & Electrical Engineering, No. 999, Qingnian Road, Hangzhou, Zhejiang 310000, China
2 shared publications
Latest Publications
- Open Access
- 1 January 2026
- Published by Institution of Engineering and Technology (IET) in IET Computer Vision
- Vol. 20 (1)
MMCATrack: Multi‐Modal Channel Attention Tracker
Most existing Transformer‐based visual object tracking methods rely exclusively on the feature map from the last encoder layer for object prediction, thereby overlooking the rich information contained in shallow and intermediate layer feature maps. This limitation reduces the representational capacity of the model. Moreover, current multi‐modal tracking frameworks typically construct multi‐modal features through simple concatenation, which fails to adequately account for the differential contributions of individual modalities to the final prediction task. As a result, these approaches exhibit an insufficient ability to express key features within the multi‐modal representation. To address the aforementioned issues, this paper proposes a multi‐modal channel attention tracking algorithm, where a multi‐modal channel attention block is incorporated for the purpose of enhancing the representation ability of the key features within the multi‐modal features. Specifically, the multi‐modal channel attention block first aggregates multi‐modal information from the multi‐layer feature maps of the encoder through cross layer cascading and then applies channel attention mechanism to dynamically calibrate the channel weights in the generated multi‐modal features, thereby enhancing the representation of key features. In addition, this article proposes a new regression loss function to improve localisation accuracy. Finally, abundant experiments conducted on five benchmarks including GOT‐10K, TrackingNet, TNL2K, VisEvent and RGBT234 have verified the effectiveness of our theory.
- 1 January 2026
- Published by Institution of Engineering and Technology (IET) in IET Computer Vision
- Vol. 20 (1)
Open AccessMMCATrack: Multi‐Modal Channel Attention Tracker
Most existing Transformer‐based visual object tracking methods rely exclusively on the feature map from the last encoder layer for object prediction, thereby overlooking the rich information contained in shallow and intermediate layer feature maps. This limitation reduces the representational capacity of the model. Moreover, current multi‐modal tracking frameworks typically construct multi‐modal features through simple concatenation, which fails to adequately account for the differential contributions of individual modalities to the final prediction task. As a result, these approaches exhibit an insufficient ability to express key features within the multi‐modal representation. To address the aforementioned issues, this paper proposes a multi‐modal channel attention tracking algorithm, where a multi‐modal channel attention block is incorporated for the purpose of enhancing the representation ability of the key features within the multi‐modal features. Specifically, the multi‐modal channel attention block first aggregates multi‐modal information from the multi‐layer feature maps of the encoder through cross layer cascading and then applies channel attention mechanism to dynamically calibrate the channel weights in the generated multi‐modal features, thereby enhancing the representation of key features. In addition, this article proposes a new regression loss function to improve localisation accuracy. Finally, abundant experiments conducted on five benchmarks including GOT‐10K, TrackingNet, TNL2K, VisEvent and RGBT234 have verified the effectiveness of our theory.
- Open Access
- 24 October 2025
- Published by Springer Nature in Scientific Reports
- Vol. 15 (1) , 1-12
Object state optimization algorithm based on Bayesian random sampling for visual object tracking
From the perspective of object state modeling, visual object tracking can be regarded as a unified process that combines object state estimation and object localization. In this framework, state estimation refers to predicting the complete state vector of the object–such as its position, scale, and motion dynamics–while localization specifically denotes identifying the object’s spatial position within the image, typically in the form of bounding box coordinates. Traditional optimization-based methods for state estimation often suffer from getting trapped in local optima, primarily due to the non-convexity of the objective function and the algorithm’s sensitivity to initialization. To address these issues, this research proposes an object state optimization algorithm based on Bayesian random sampling for visual object tracking. Firstly, a dense sampling method is introduced to mitigate the problem of local optima. Secondly, a hybrid model that merges Bayesian random sampling and gradient ascent is proposed to refine the bounding box, successfully alleviating convergence instability. Finally, our experimental results show that the proposed algorithm significantly improves tracking performance on multiple datasets, validating its efficiency and applicability in object state estimation tasks.
- 24 October 2025
- Published by Springer Nature in Scientific Reports
- Vol. 15 (1) , 1-12
Open AccessObject state optimization algorithm based on Bayesian random sampling for visual object tracking
From the perspective of object state modeling, visual object tracking can be regarded as a unified process that combines object state estimation and object localization. In this framework, state estimation refers to predicting the complete state vector of the object–such as its position, scale, and motion dynamics–while localization specifically denotes identifying the object’s spatial position within the image, typically in the form of bounding box coordinates. Traditional optimization-based methods for state estimation often suffer from getting trapped in local optima, primarily due to the non-convexity of the objective function and the algorithm’s sensitivity to initialization. To address these issues, this research proposes an object state optimization algorithm based on Bayesian random sampling for visual object tracking. Firstly, a dense sampling method is introduced to mitigate the problem of local optima. Secondly, a hybrid model that merges Bayesian random sampling and gradient ascent is proposed to refine the bounding box, successfully alleviating convergence instability. Finally, our experimental results show that the proposed algorithm significantly improves tracking performance on multiple datasets, validating its efficiency and applicability in object state estimation tasks.
- Open Access
- 1 January 2025
- Published by Wiley in International Journal of Intelligent Systems
- Vol. 2025 (1)
A Protecting‐Privacy Path Query Supporting Semantic‐Based Multikeyword Search Over Ciphertext Graph Data in Cloud Computing
With the rapid development of information technology and intelligence, there are more and more usage scenarios of graph data. Path queries have always been a hot topic of research for scholars. There are already many mature methods and ideas in the study of the path query on the plaintext graph. For the path query on graph data in the case of cloud outsourcing, it is necessary to consider both the construction of query algorithms and the protection of privacy information. The processing of graph privacy information through encryption, and then outsourcing to the cloud platform, is a common measure. The semantic‐based path query supporting multikeyword search is an extended path query method, which can improve the query function. For users, it is very troublesome to access and process the encrypted graph data. In this article, we propose a protecting‐privacy semantic‐based multikeyword path query scheme on the ciphertext graph (PSMP). Firstly, based on the principle of searchable encryption and vector operation, a secure index is constructed, and then the cloud server uses the secure index to implement path queries. This article demonstrates its security through formal analysis and verifies its effectiveness through experimental comparison and analysis. The work of this article has a certain promoting effect on the query processing and analysis of ciphertext graph data.
- 1 January 2025
- Published by Wiley in International Journal of Intelligent Systems
- Vol. 2025 (1)
Open AccessA Protecting‐Privacy Path Query Supporting Semantic‐Based Multikeyword Search Over Ciphertext Graph Data in Cloud Computing
With the rapid development of information technology and intelligence, there are more and more usage scenarios of graph data. Path queries have always been a hot topic of research for scholars. There are already many mature methods and ideas in the study of the path query on the plaintext graph. For the path query on graph data in the case of cloud outsourcing, it is necessary to consider both the construction of query algorithms and the protection of privacy information. The processing of graph privacy information through encryption, and then outsourcing to the cloud platform, is a common measure. The semantic‐based path query supporting multikeyword search is an extended path query method, which can improve the query function. For users, it is very troublesome to access and process the encrypted graph data. In this article, we propose a protecting‐privacy semantic‐based multikeyword path query scheme on the ciphertext graph (PSMP). Firstly, based on the principle of searchable encryption and vector operation, a secure index is constructed, and then the cloud server uses the secure index to implement path queries. This article demonstrates its security through formal analysis and verifies its effectiveness through experimental comparison and analysis. The work of this article has a certain promoting effect on the query processing and analysis of ciphertext graph data.
- Open Access
- 15 December 2023
- Published by University of Slavonski Brod in Tehnicki vjesnik - Technical Gazette
- Vol. 30 (6) , 1704-1716
Enabling Access Control for Encrypted Multi-Dimensional Data in Cloud Computing through Range Search
With the growing popularity of cloud computing, data owners are increasingly opting to outsource their data to cloud servers due to the numerous benefits it offers. However, this outsourcing raises concerns about data privacy since the data stored on...
- 15 December 2023
- Published by University of Slavonski Brod in Tehnicki vjesnik - Technical Gazette
- Vol. 30 (6) , 1704-1716
Open AccessEnabling Access Control for Encrypted Multi-Dimensional Data in Cloud Computing through Range Search
With the growing popularity of cloud computing, data owners are increasingly opting to outsource their data to cloud servers due to the numerous benefits it offers. However, this outsourcing raises concerns about data privacy since the data stored on...
- 10 August 2023
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 83 (8) , 23047-23065
Robust object tracking based on power-law probability map and ridge regression
- 10 August 2023
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 83 (8) , 23047-23065
Robust object tracking based on power-law probability map and ridge regression
- 1 December 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1132-1138
Attention-Based Convolutional Neural Network and Bidirectional Gated Recurrent Unit for Human Activity Recognition
Using the data collected by the sensors embedded in smartphones for human activity recognition has become an important research direction. In order to overcome the problem of handcrafted feature extraction in traditional machine learning, we propose an attention-based model which combines convolutional neural network (CNN) and bidirectional gated recurrent unit (BiGRU) for human activity recognition. This model can extract local features of the original data by CNN, and construct the extracted features into a time series as the input of BiGRU, the BiGRU can capture the temporal dependencies and extract the features from the input, by introducing the attention mechanism, the influence of key input can be strengthened. The features are classified by the Softmax function. The recognition accuracy rate on the Wireless Sensor Data Mining (WISDM) dataset reaches 98.5%, which is improved compared with the state-of-the-art methods.
- 1 December 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1132-1138
Attention-Based Convolutional Neural Network and Bidirectional Gated Recurrent Unit for Human Activity Recognition
Using the data collected by the sensors embedded in smartphones for human activity recognition has become an important research direction. In order to overcome the problem of handcrafted feature extraction in traditional machine learning, we propose an attention-based model which combines convolutional neural network (CNN) and bidirectional gated recurrent unit (BiGRU) for human activity recognition. This model can extract local features of the original data by CNN, and construct the extracted features into a time series as the input of BiGRU, the BiGRU can capture the temporal dependencies and extract the features from the input, by introducing the attention mechanism, the influence of key input can be strengthened. The features are classified by the Softmax function. The recognition accuracy rate on the Wireless Sensor Data Mining (WISDM) dataset reaches 98.5%, which is improved compared with the state-of-the-art methods.
- Open Access
- 14 June 2020
- Published by University of Slavonski Brod in Tehnicki vjesnik - Technical Gazette
- Vol. 27 (3) , 923-932
Query with Assumptions for Probabilistic Relational Databases
Users may have prior knowledge about a probabilistic database. They prefer to query over a probabilistic database on their prior knowledge which cannot be written as component clauses of conventional SQL queries. A naive approach is to...
- 14 June 2020
- Published by University of Slavonski Brod in Tehnicki vjesnik - Technical Gazette
- Vol. 27 (3) , 923-932
Open AccessQuery with Assumptions for Probabilistic Relational Databases
Users may have prior knowledge about a probabilistic database. They prefer to query over a probabilistic database on their prior knowledge which cannot be written as component clauses of conventional SQL queries. A naive approach is to...
- Open Access
- 16 September 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 7, 133716-133724
Secure and Efficient Adjacency Search Supporting Synonym Query on Encrypted Graph in the Cloud
With the popularity of cloud computing, cloud outsourcing has aroused great concern in academia. A lot of graph data are outsourced to the cloud for saving cost. As the cloud server may be not entirely reliable, the outsourced graph data are usually encrypted before sending to the cloud for the security considerations. Adjacency search is a basic operation, and many other operations can be performed based on the adjacency search. Adjacency search supporting synonym query is a more meaningful operation which can make a query user improve the scope of the query. Due to the graph data being encrypted in the cloud, adjacency search supporting synonym query becomes a very challenging task. In this paper, we propose an efficient solution to perform privacy-assured adjacency search supporting synonym query on the encrypted graph in the cloud (PASQ). Our work utilizes a stemming algorithm and an encryption mechanism to perform adjacency search. A query user can get a wider range of search results by this solution, and the privacy of query will not be disclosed. We present a design scheme for the outsourced graph and analyze the security. We demonstrate the efficiency of our scheme by the experiment results on a real graph data.
- 16 September 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 7, 133716-133724
Open AccessSecure and Efficient Adjacency Search Supporting Synonym Query on Encrypted Graph in the Cloud
With the popularity of cloud computing, cloud outsourcing has aroused great concern in academia. A lot of graph data are outsourced to the cloud for saving cost. As the cloud server may be not entirely reliable, the outsourced graph data are usually encrypted before sending to the cloud for the security considerations. Adjacency search is a basic operation, and many other operations can be performed based on the adjacency search. Adjacency search supporting synonym query is a more meaningful operation which can make a query user improve the scope of the query. Due to the graph data being encrypted in the cloud, adjacency search supporting synonym query becomes a very challenging task. In this paper, we propose an efficient solution to perform privacy-assured adjacency search supporting synonym query on the encrypted graph in the cloud (PASQ). Our work utilizes a stemming algorithm and an encryption mechanism to perform adjacency search. A query user can get a wider range of search results by this solution, and the privacy of query will not be disclosed. We present a design scheme for the outsourced graph and analyze the security. We demonstrate the efficiency of our scheme by the experiment results on a real graph data.
- 16 September 2019
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 79 (1-2) , 785-804
Robust object tracking based on ridge regression and multi-scale local sparse coding
- 16 September 2019
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 79 (1-2) , 785-804
Robust object tracking based on ridge regression and multi-scale local sparse coding
- 13 August 2019
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 78 (24) , 34725-34744
Visual object tracking with discriminative correlation filtering and hybrid color feature
- 13 August 2019
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 78 (24) , 34725-34744
Visual object tracking with discriminative correlation filtering and hybrid color feature