Man Chen
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
Man Chen has a total of 46 co-authors. Here are the top co-authors by number of shared publications:
- Zhisong Pan
Army Engineering University of PLA
8 shared publications Huaping Zhou
4 shared publications- Yifei Xie
Command and Control Engineering College, Army Engineering University, Nanjing, 210007, Jiangsu, China
4 shared publications - Enping Chen
School of Electrical and Information Engineering, Changsha University of Science and Technology, Changsha 410114, China
3 shared publications - Kelei Sun
School of Computer Science and Engineering, Anhui University of Science and Technology, Huainan, China
3 shared publications - Yao Zhang
College of Command and Control Engineering, Army Engineering University of PLA, Nanjing 210007, China
3 shared publications
Latest Publications
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Consumer Electronics
- Vol. 72 (4) , 10440-10452
DIVERSE-DAH: A Dual-Attention Heterogeneous Graph for Diversity-Aware Recommendation
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Consumer Electronics
- Vol. 72 (4) , 10440-10452
DIVERSE-DAH: A Dual-Attention Heterogeneous Graph for Diversity-Aware Recommendation
- 22 May 2026
- Published by Elsevier in Applied Soft Computing
- Vol. 201
Robust semi-supervised SAR ship detection under joint constraints of annotation scale and label quality
- 22 May 2026
- Published by Elsevier in Applied Soft Computing
- Vol. 201
Robust semi-supervised SAR ship detection under joint constraints of annotation scale and label quality
- 2 December 2025
- Published by Elsevier in Neural Networks
- Vol. 196, 108369
Full-spectrum prompt tuning with sparse MoE for open-set recognition
- 2 December 2025
- Published by Elsevier in Neural Networks
- Vol. 196, 108369
Full-spectrum prompt tuning with sparse MoE for open-set recognition
- Open Access
- 3 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 18 (19391404) , 26573-26585
CGFRNet: A Context-Guided and Feature Refined Network for SAR Image Ship Detection
Synthetic Aperture Radar (SAR)-based remotely sensed imagery plays a crucial role in both civil and military domains. However, complex background interference, dense distribution, and different ship sizes are the current interrelated and unsolvable challenges. To this end, we propose a context-guided and feature refinement network (CGFRNet) for ship detection in SAR images. First, a feature refinement module (FRM) is designed to capture spatial contextual information of targets at different scales while suppressing easily confused background interference. Second, the context-guided attention module (CGAM) is introduced to fully integrate the shallow feature mapping with the deeper layers further to enhance the focusing and extraction of key features of the ship. Third, due to the dense distribution of most small ships in SAR images, the separation enhanced attention module (SEAM) is incorporated into the detection head to learn the dependencies between features and dynamically adjust the attention points in the feature maps to enhance the model performance in dense target scenes. Experimental results on HRSID, SSDD, and LS-SSDD-v1.0 datasets show that the proposed method can significantly improve the detection accuracy of multi-scale targets in complex scenes and achieve a more advanced performance.
- 3 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 18 (19391404) , 26573-26585
Open AccessCGFRNet: A Context-Guided and Feature Refined Network for SAR Image Ship Detection
Synthetic Aperture Radar (SAR)-based remotely sensed imagery plays a crucial role in both civil and military domains. However, complex background interference, dense distribution, and different ship sizes are the current interrelated and unsolvable challenges. To this end, we propose a context-guided and feature refinement network (CGFRNet) for ship detection in SAR images. First, a feature refinement module (FRM) is designed to capture spatial contextual information of targets at different scales while suppressing easily confused background interference. Second, the context-guided attention module (CGAM) is introduced to fully integrate the shallow feature mapping with the deeper layers further to enhance the focusing and extraction of key features of the ship. Third, due to the dense distribution of most small ships in SAR images, the separation enhanced attention module (SEAM) is incorporated into the detection head to learn the dependencies between features and dynamically adjust the attention points in the feature maps to enhance the model performance in dense target scenes. Experimental results on HRSID, SSDD, and LS-SSDD-v1.0 datasets show that the proposed method can significantly improve the detection accuracy of multi-scale targets in complex scenes and achieve a more advanced performance.
- Open Access
- 14 March 2025
- Published by Association for Computing Machinery (ACM)
- p. 229-234
A Flow-Based Cross Projection Framework for Unsupervised Text Style Transfer
- 14 March 2025
- Published by Association for Computing Machinery (ACM)
- p. 229-234
Open AccessA Flow-Based Cross Projection Framework for Unsupervised Text Style Transfer
- Open Access
- 29 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 17 (19391404) , 15685-15701
Scale-Mixing Enhancement and Dual Consistency Guidance for End-to-End Semisupervised Ship Detection in SAR Images
The object detection of synthetic aperture radar (SAR) ships holds significant promise for water traffic monitoring, ship search and rescue, and maritime warning tasks. Regrettably, most of the existing SAR ship object detection is limited to the fully supervised paradigm, exhibiting a firm reliance on data labels, and inherent challenges such as multi-scale ship feature disparities and indistinct small-sized ships also make SAR ship detection difficult. To address these, we propose a scale mixing enhanced and dual consistency guided semi-supervised object detection (SMDC-SSOD) method. Specifically, this method is based on the teacher-student framework (TSF) and primarily comprises three core components: cross-scale feature mixing (CSFM) scheme, scale change consistency guidance (SCCG) strategy, and proposal consistency guidance (PCG) strategy, which can efficiently conduct end-to-end semi-supervised learning from limited data labeling, achieving low-cost and high-performance ship perception. CSFM scheme includes inter-pyramid and intra-pyramid feature cross-scale mixings, which can improve the network's adaptability for multi-scale ship characteristics and increase focus on small-sized ships. SCCG strategy leverages variations in confidence scores at different scales to select valuable pseudo-labels, providing more precise guidance for the student network. PCG strategy further reflects the positioning quality of pseudo-labels through the proposal consistency generated by the student network, guiding it to make high-quality predictions. The experimental results on the publicly available HRSID, BBox-SSDD, and SAR-Ship-Dataset demonstrate that SMDC-SSOD can accurately detect SAR ships with an extremely low data annotation rate (below 10%) and achieve optimal detection performance compared to state-of-the-art methods.
- 29 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 17 (19391404) , 15685-15701
Open AccessScale-Mixing Enhancement and Dual Consistency Guidance for End-to-End Semisupervised Ship Detection in SAR Images
The object detection of synthetic aperture radar (SAR) ships holds significant promise for water traffic monitoring, ship search and rescue, and maritime warning tasks. Regrettably, most of the existing SAR ship object detection is limited to the fully supervised paradigm, exhibiting a firm reliance on data labels, and inherent challenges such as multi-scale ship feature disparities and indistinct small-sized ships also make SAR ship detection difficult. To address these, we propose a scale mixing enhanced and dual consistency guided semi-supervised object detection (SMDC-SSOD) method. Specifically, this method is based on the teacher-student framework (TSF) and primarily comprises three core components: cross-scale feature mixing (CSFM) scheme, scale change consistency guidance (SCCG) strategy, and proposal consistency guidance (PCG) strategy, which can efficiently conduct end-to-end semi-supervised learning from limited data labeling, achieving low-cost and high-performance ship perception. CSFM scheme includes inter-pyramid and intra-pyramid feature cross-scale mixings, which can improve the network's adaptability for multi-scale ship characteristics and increase focus on small-sized ships. SCCG strategy leverages variations in confidence scores at different scales to select valuable pseudo-labels, providing more precise guidance for the student network. PCG strategy further reflects the positioning quality of pseudo-labels through the proposal consistency generated by the student network, guiding it to make high-quality predictions. The experimental results on the publicly available HRSID, BBox-SSDD, and SAR-Ship-Dataset demonstrate that SMDC-SSOD can accurately detect SAR ships with an extremely low data annotation rate (below 10%) and achieve optimal detection performance compared to state-of-the-art methods.
- Open Access
- 2 July 2024
- Published by Springer Nature in Complex & Intelligent Systems
- Vol. 10 (5) , 6965-6982
Model inductive bias enhanced deep reinforcement learning for robot navigation in crowded environments
Navigating mobile robots in crowded environments poses a significant challenge and is essential for the coexistence of robots and humans in future intelligent societies. As a pragmatic data-driven approach, deep reinforcement learning (DRL) holds promise for addressing this challenge. However, current DRL-based navigation methods have possible improvements in understanding agent interactions, feedback mechanism design, and decision foresight in dynamic environments. This paper introduces the model inductive bias enhanced deep reinforcement learning (MIBE-DRL) method, drawing inspiration from a fusion of data-driven and model-driven techniques. MIBE-DRL extensively incorporates model inductive bias into the deep reinforcement learning framework, enhancing the efficiency and safety of robot navigation. The proposed approach entails a multi-interaction network featuring three modules designed to comprehensively understand potential agent interactions in dynamic environments. The pedestrian interaction module can model interactions among humans, while the temporal and spatial interaction modules consider agent interactions in both temporal and spatial dimensions. Additionally, the paper constructs a reward system that fully accounts for the robot’s direction and position factors. This system's directional and positional reward functions are built based on artificial potential fields (APF) and navigation rules, respectively, which can provide reasoned evaluations for the robot's motion direction and position during training, enabling it to receive comprehensive feedback. Furthermore, the incorporation of Monte-Carlo tree search (MCTS) facilitates the development of a foresighted action strategy, enabling robots to execute actions with long-term planning considerations. Experimental results demonstrate that integrating model inductive bias significantly enhances the navigation performance of MIBE-DRL. Compared to state-of-the-art methods, MIBE-DRL achieves the highest success rate in crowded environments and demonstrates advantages in navigation time and maintaining a safe social distance from humans.
- 2 July 2024
- Published by Springer Nature in Complex & Intelligent Systems
- Vol. 10 (5) , 6965-6982
Open AccessModel inductive bias enhanced deep reinforcement learning for robot navigation in crowded environments
Navigating mobile robots in crowded environments poses a significant challenge and is essential for the coexistence of robots and humans in future intelligent societies. As a pragmatic data-driven approach, deep reinforcement learning (DRL) holds promise for addressing this challenge. However, current DRL-based navigation methods have possible improvements in understanding agent interactions, feedback mechanism design, and decision foresight in dynamic environments. This paper introduces the model inductive bias enhanced deep reinforcement learning (MIBE-DRL) method, drawing inspiration from a fusion of data-driven and model-driven techniques. MIBE-DRL extensively incorporates model inductive bias into the deep reinforcement learning framework, enhancing the efficiency and safety of robot navigation. The proposed approach entails a multi-interaction network featuring three modules designed to comprehensively understand potential agent interactions in dynamic environments. The pedestrian interaction module can model interactions among humans, while the temporal and spatial interaction modules consider agent interactions in both temporal and spatial dimensions. Additionally, the paper constructs a reward system that fully accounts for the robot’s direction and position factors. This system's directional and positional reward functions are built based on artificial potential fields (APF) and navigation rules, respectively, which can provide reasoned evaluations for the robot's motion direction and position during training, enabling it to receive comprehensive feedback. Furthermore, the incorporation of Monte-Carlo tree search (MCTS) facilitates the development of a foresighted action strategy, enabling robots to execute actions with long-term planning considerations. Experimental results demonstrate that integrating model inductive bias significantly enhances the navigation performance of MIBE-DRL. Compared to state-of-the-art methods, MIBE-DRL achieves the highest success rate in crowded environments and demonstrates advantages in navigation time and maintaining a safe social distance from humans.