Jiaju Wu
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
Jiaju Wu has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Hanrui Wu
Jinan University
1 shared publication Hong Zhu
School of Mathematical Sciences, Jiangsu University, Zhenjiang, China
1 shared publication- Jianguo Yan
Northwestern Polytechnical University
1 shared publication - Lingwei Dang
South China University of Technology, Guangzhou, China
1 shared publication - Michael KwokPo Ng
Department of Mathematics, Hong Kong Baptist University, Kowloon Tong, Hong Kong
1 shared publication - Qingyao Wu
South China University of Technology, Guangzhou, China
1 shared publication
Latest Publications
- 7 April 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 323
ExPart3D : Exclusive part aggregation for zero-shot point cloud learning
- 7 April 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 323
ExPart3D : Exclusive part aggregation for zero-shot point cloud learning
- 8 July 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 30 (10577149) , 6364-6376
Heterogeneous Domain Adaptation by Information Capturing and Distribution Matching
Heterogeneous domain adaptation (HDA) is a challenging problem because of the different feature representations in the source and target domains. Most HDA methods search for mapping matrices from the source and target domains to discover latent features for learning. However, these methods barely consider the reconstruction error to measure the information loss during the mapping procedure. In this paper, we propose to jointly capture the information and match the source and target domain distributions in the latent feature space. In the learning model, we propose to minimize the reconstruction loss between the original and reconstructed representations to preserve information during transformation and reduce the Maximum Mean Discrepancy between the source and target domains to align their distributions. The resulting minimization problem involves two projection variables with orthogonal constraints that can be solved by the generalized gradient flow method, which can preserve orthogonal constraints in the computational procedure. We conduct extensive experiments on several image classification datasets to demonstrate that the effectiveness and efficiency of the proposed method are better than those of state-of-the-art HDA methods.
- 8 July 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 30 (10577149) , 6364-6376
Heterogeneous Domain Adaptation by Information Capturing and Distribution Matching
Heterogeneous domain adaptation (HDA) is a challenging problem because of the different feature representations in the source and target domains. Most HDA methods search for mapping matrices from the source and target domains to discover latent features for learning. However, these methods barely consider the reconstruction error to measure the information loss during the mapping procedure. In this paper, we propose to jointly capture the information and match the source and target domain distributions in the latent feature space. In the learning model, we propose to minimize the reconstruction loss between the original and reconstructed representations to preserve information during transformation and reduce the Maximum Mean Discrepancy between the source and target domains to align their distributions. The resulting minimization problem involves two projection variables with orthogonal constraints that can be solved by the generalized gradient flow method, which can preserve orthogonal constraints in the computational procedure. We conduct extensive experiments on several image classification datasets to demonstrate that the effectiveness and efficiency of the proposed method are better than those of state-of-the-art HDA methods.
- 1 October 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 46, 1665-1669
Object tracking algorithm for UAV autonomous Aerial Refueling
In order to improve the docking success rate in Automated Aerial Refueling (AAR), it is important to identify the receiver aircraft's receptacle for boom receptacle refueling (BRR). Meanshift tracking algorithm only considers the H component color statistics of the target area, lacking spatial information, could easily lead to inaccurate tracking. Besides, Meanshift tracking algorithm could easily lost target under occlusion conditions. To handle these situations, this paper proposes an improved Meanshift tracking algorithm based on color fusion and kernel function combined with Kalman filter (IMS_KF). In view of lacking color component, use RGB linear fusion. In view of lacking spatial information, define the kernel function by setting different weight to pixels, on the basis of the distance from the center point of target to the current point. In view of occlusion conditions, use Kalman Filter algorithm to estimate the location of moving targets. Meanshift tracking results will determine whether use Kalman forecasting. We implemented this algorithm on F-16 simulation experiment platform and the results reveal that our method meets industrial real-time requirements and has a better tracking robustness under a complex environment.
- 1 October 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 46, 1665-1669
Object tracking algorithm for UAV autonomous Aerial Refueling
In order to improve the docking success rate in Automated Aerial Refueling (AAR), it is important to identify the receiver aircraft's receptacle for boom receptacle refueling (BRR). Meanshift tracking algorithm only considers the H component color statistics of the target area, lacking spatial information, could easily lead to inaccurate tracking. Besides, Meanshift tracking algorithm could easily lost target under occlusion conditions. To handle these situations, this paper proposes an improved Meanshift tracking algorithm based on color fusion and kernel function combined with Kalman filter (IMS_KF). In view of lacking color component, use RGB linear fusion. In view of lacking spatial information, define the kernel function by setting different weight to pixels, on the basis of the distance from the center point of target to the current point. In view of occlusion conditions, use Kalman Filter algorithm to estimate the location of moving targets. Meanshift tracking results will determine whether use Kalman forecasting. We implemented this algorithm on F-16 simulation experiment platform and the results reveal that our method meets industrial real-time requirements and has a better tracking robustness under a complex environment.