College of Computer Science and Software Engineering, Hohai University, China
Zizun Wei is a Master’s student at the College of Computer Science and Software Engineering, Hohai University, Nanjing, China. At 24 years old, he is highly passionate about the intersection of artificial intelligence, computer vision, and water resources. Zizun has demonstrated a strong academic foundation, having completed his Bachelor’s degree in Network Engineering at Heilongjiang University, Harbin. His research interests include object detection, few-shot learning, and applying AI technologies to real-world challenges, particularly in environmental monitoring. Zizun has contributed significantly to scientific literature, with multiple published papers in respected journals and conferences. Additionally, his innovative work in the field has led to a patent in semantic segmentation for water body extraction. His work is at the forefront of AI’s application to environmental science, particularly focusing on river debris detection and crack segmentation in earth dams. Zizun’s drive to combine AI with practical solutions positions him as a promising researcher in his field.
Profile
Google Scholar
Education
Zizun Wei is currently pursuing a Master’s degree in Computer Science and Technology at Hohai University, Nanjing, from September 2022 to June 2025. His academic journey began with a Bachelor’s degree in Network Engineering from Heilongjiang University, Harbin, where he studied from September 2018 to June 2022. During his Bachelor’s, Zizun laid the groundwork for his research interests in artificial intelligence, network systems, and computer vision. At Hohai University, he is expanding his knowledge in computer science, with a focus on object detection, machine learning, and their application to environmental challenges, such as water resource management. Zizun has excelled in both practical and theoretical aspects of his education, participating in multiple research projects while refining his skills in data analysis and AI modeling. His academic path reflects a dedication to advancing technology in ways that contribute to society, particularly in the areas of environmental protection and resource management.
Experience
Zizun Wei has gained diverse research experience throughout his academic career, starting with his involvement in salient object detection during his Bachelor’s degree, where he worked on feature fusion and pixel loss weighting methods. His Master’s research has broadened to focus on few-shot object detection and river floating debris detection. He proposed a meta-feature extraction approach for few-shot object detection and worked on enhancing algorithms using generative fitting for real-world data. Zizun is also exploring text knowledge embedding techniques to enhance the performance of object detection models. His work on river floating debris detection aimed at improving feature enhancement methods, with a view to addressing environmental challenges in water bodies. Additionally, Zizun co-authored papers in top-tier journals and conferences, as well as a patent on semantic segmentation methods for water body extraction. His research not only advances the field of computer vision but also addresses practical environmental concerns, reflecting his interdisciplinary approach.
Research Focus
Zizun Wei’s research primarily revolves around computer vision and artificial intelligence, with a particular focus on object detection and few-shot learning. His work in object detection has been aimed at improving algorithms for complex, real-world applications such as detecting debris in rivers and cracks in earth dams. He is developing few-shot learning techniques that mimic human cognition and leverage transfer learning for more efficient detection in environments with limited labeled data. Zizun’s innovations include the use of meta-feature extraction and generative fitting methods to enhance detection performance, particularly in the context of environmental and water resources. He is also exploring the integration of text knowledge embeddings to further advance the performance of detection models. With an overarching goal of contributing to sustainable water resource management, his work combines cutting-edge AI techniques with real-world applications that can make significant environmental impacts.
Publication Top Notes
- L. Zhang, Z. Wei, Y. Shao, Z. Chen, Z. Luo, and Y. Dou, “A context feature enhancement and adaptive weighted fusion network for river floating debris detection,” Engineering Applications of Artificial Intelligence, Mar. 2025.
- L. Zhang, Z. Wei, and P. Jin, “DAMFE-Net: A Few-shot Crack Segmentation Model Based on Transfer Learning for Earth Dams,” IEEE 15th International Conference on Software Engineering and Service Science (ICSESS), 2024.
- Z. Wei and G. Zhu, “A Salient Object Detection Method Combining Multi-Scale Feature Fusion and Pixel Loss Weighting,” Journal of Natural Science of Heilongjiang University, 2022.
- G. Zhu, Z. Wei, and F. Lin, “An Object Detection Method Combining Multi-Level Feature Fusion and Region Channel Attention,” IEEE Access, 2021.
- Patent: “A Lightweight Dual-Prediction Branch Semantic Segmentation Deep Learning Method and System for Water Body Extraction.” CN117911701A. Co-inventors: Zhang Lili, Wei Zizun, Lu Yushi, Wang Huibin, Chen Jun, Chen Zhe. Publication Date: April 19, 2024.