Jun Wu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Jun Wu has a total of 20 co-authors. Here are the top co-authors by number of shared publications:
- Li Yang
School of Computer Science, Hubei University of Education, Wuhan, China
3 shared publications - Gan Zuo
IBM consulting (Wuhan), Wuhan, China
2 shared publications - Pengfei Zhan
School of Computer Science, Hubei University of Technology, Wuhan, China
2 shared publications - Shuai Guo
School of Computer Science, Hubei University of Technology, Wuhan, China
2 shared publications - Yu Chen
School of Computer Science, Hubei University of Technology, Wuhan, China
2 shared publications - Yu Zheng
School of Computer Science, Hubei University of Technology, Wuhan, China
2 shared publications
Latest Publications
- 5 May 2026
- Published by Springer Nature in The Journal of Supercomputing
- Vol. 82 (7) , 394
MSS-UNet: Mamba-based multi-directional selective scanning for medical image segmentation using dermoscopic and multimodal datasets
- 5 May 2026
- Published by Springer Nature in The Journal of Supercomputing
- Vol. 82 (7) , 394
MSS-UNet: Mamba-based multi-directional selective scanning for medical image segmentation using dermoscopic and multimodal datasets
- 9 March 2026
- Published by Springer Nature in Multimedia Systems
- Vol. 32 (3) , 159
Auxiliary enhanced adaptive entropy-guided multi-granularity networks for multimodal emotion recognition in conversations
- 9 March 2026
- Published by Springer Nature in Multimedia Systems
- Vol. 32 (3) , 159
Auxiliary enhanced adaptive entropy-guided multi-granularity networks for multimodal emotion recognition in conversations
- Open Access
- 20 October 2025
- Published by Springer Nature in Research Square
MSS-UNet : Mamba-Based Multi-directional Selective Scanning for Medical Image Segmentation
Medical image segmentation is a critical task in the field of computer vision, aimed at segmenting specific organs or lesions from input medical images. In current research, multiple deep learning architectures including Convolutional Neural Networks (CNNs) and Transformers are widely employed to tackle this challenge. Nevertheless, CNN-based methods remain constrained in establishing long-range dependencies, whereas Transformer architectures, despite demonstrating superior modeling capacities, are hindered by quadratic computational complexity. To overcome these dual limitations, this study introduces MSS-UNet, a novel architecture integrating State-Space Models (SSMs) that achieves linear computational scaling while maintaining robust long-range modeling capabilities, thereby effectively mitigating the respective shortcomings of CNN and Transformer approaches. We use the U-Net architecture as the overall framework and stack four MSS blocks in the encoder part to capture broad contextual information. In the skip connection part, we introduce the DCA module to enhance the fusion of low-level and high-level features. Extensive experiments were conducted on public benchmark datasets including ISIC2017 and ISIC2018. The experimental results demonstrate that the proposed MSSUNet achieves strong performance in skin lesion segmentation tasks, exhibiting promising effectiveness in medical image analysis.The codes are available at https://github.com/mmm587/MSS-UNet
- 20 October 2025
- Published by Springer Nature in Research Square
Open AccessMSS-UNet : Mamba-Based Multi-directional Selective Scanning for Medical Image Segmentation
Medical image segmentation is a critical task in the field of computer vision, aimed at segmenting specific organs or lesions from input medical images. In current research, multiple deep learning architectures including Convolutional Neural Networks (CNNs) and Transformers are widely employed to tackle this challenge. Nevertheless, CNN-based methods remain constrained in establishing long-range dependencies, whereas Transformer architectures, despite demonstrating superior modeling capacities, are hindered by quadratic computational complexity. To overcome these dual limitations, this study introduces MSS-UNet, a novel architecture integrating State-Space Models (SSMs) that achieves linear computational scaling while maintaining robust long-range modeling capabilities, thereby effectively mitigating the respective shortcomings of CNN and Transformer approaches. We use the U-Net architecture as the overall framework and stack four MSS blocks in the encoder part to capture broad contextual information. In the skip connection part, we introduce the DCA module to enhance the fusion of low-level and high-level features. Extensive experiments were conducted on public benchmark datasets including ISIC2017 and ISIC2018. The experimental results demonstrate that the proposed MSSUNet achieves strong performance in skin lesion segmentation tasks, exhibiting promising effectiveness in medical image analysis.The codes are available at https://github.com/mmm587/MSS-UNet
- 4 October 2024
- Published by Springer Nature in Journal of Intelligent Information Systems
- Vol. 63 (2) , 375-394
MLGAT: multi-layer graph attention networks for multimodal emotion recognition in conversations
- 4 October 2024
- Published by Springer Nature in Journal of Intelligent Information Systems
- Vol. 63 (2) , 375-394
MLGAT: multi-layer graph attention networks for multimodal emotion recognition in conversations
- 17 June 2024
- Published by Springer Nature in The Visual Computer
- Vol. 41 (3) , 2029-2046
Malleable pruning meets more scaled wide-area of attention model for real-time crack detection
- 17 June 2024
- Published by Springer Nature in The Visual Computer
- Vol. 41 (3) , 2029-2046
Malleable pruning meets more scaled wide-area of attention model for real-time crack detection