Prof. Zhisheng Huang
Vrije Universiteit Amsterdam, The Netherlands
Professor Zhisheng Huang is a tenured senior researcher at Computer Science Department of VU University Amsterdam, the Netherlands. He is also a full professor at School of Computer Science and Engineering, Wuhan University of Science and Technology, China. His research interests include Semantic Web technology, knowledge graphs, and ontology engineering, Artificial Intelligence for medicine, and medical informatics. Prof. Huang has published more than 300 papers in journals/conferences/workshops, served as a member of programme/organising committee for over 200 international conferences/workshops. He is the editor-in-chief of Journal of Artificial Intelligence for Medical Sciences.
Title: Knowledge Graph Technology for Smart City
Abstract: Knowledge graphs are one of the two important technological pillars of artificial intelligence. Knowledge graph technology is used to construct large knowledge bases, integrating various data and knowledge resources in a specific domain to achieve decision support. In this report, we will introduce the basic ideas, principles, and technical methods of knowledge graphs. We will also showcase the broad application prospects of knowledge graph technology by introducing specific application cases in smart cities. We will specifically introduce some successful knowledge graph projects used in smart cities in the Netherlands and the tree hole rescue project in China.
Prof. Huafeng Li
Kunming University of Science and Technology, China
Huaifeng Li, Ph.D., is a Professor and Ph.D. Supervisor at Kunming University of Science and Technology. He is a recipient of the Outstanding Young Scientist Fund of Yunnan Province and a Young Top-notch Talent of the Yunnan Talents Support Program. His academic editorial roles include Associate Editor of IEEE Transactions on Image Processing (IEEE TIP) and IEEE Signal Processing Letters (IEEE SPL), Editorial Board Member of Information Fusion, Senior Program Committee (SPC) Member of the top international artificial intelligence conference AAAI, and Young Editorial Board Member of the Journal of Chongqing University. He also serves as a member of the Youth Working Committee of the Chinese Society of Image and Graphics (CSIG), as well as a member of the Machine Vision Technical Committee and Biometrics Technical Committee of CSIG; in addition, he holds memberships in the Technical Committee on Intelligent Services and Technical Committee on Intelligent Fusion of the Chinese Association for Artificial Intelligence (CAAI), and the Technical Committee on Pattern Recognition and Machine Intelligence of the Chinese Association of Automation (CAA). As principal investigator, he has presided over 4 projects funded by the National Natural Science Foundation of China, 1 project funded by the Outstanding Young Scientist Fund of Yunnan Province, 1 Key Basic Research Project of Yunnan Province, and 1 Youth Science Foundation Project of Yunnan Province. He has published more than 70 academic papers in high-level international journals and conferences such as CVPR, ICCV, AAAI, ACMMM, IJCV, IEEE TPAMI, IEEE TIP and IEEE TIFS. He has won the Second-Class Prize of Yunnan Provincial Natural Science Award (ranked first), obtained more than 50 authorized national invention patents, and accomplished 10 technology transfer projects.
Title: Thinking and Future Trends of Person Re-identification in Cross-modal and Weakly-supervised Scenarios
Abstract: This report focuses on reflections and future directions of person re-identification, highlighting our research progress in weakly-supervised and cross-modal scenarios, and further discussing new paradigms for person retrieval and localization. First, it introduces weakly-supervised text-to-person image retrieval and weakly-supervised infrared-visible person matching methods, which enhance retrieval performance in complex scenarios via limited annotations and cross-modal alignment. Second, it shares the use of negative samples to train domain generalization models, mitigating performance degradation caused by cross-scene and cross-device variations. Strategies to reduce pseudo-label prediction inaccuracies in clustering frameworks are also explored, including the modeling and correction mechanisms for pseudo-label uncertainty. On this basis, the report presents a unified retrieval-and-localization solution that enables accurate target person search and spatial positioning, and systematically analyzes its differences and advantages over traditional person re-identification methods.
Prof. Zhaohong Deng
Jiangnan University, China
Zhao-Hong Deng, Professor and Doctoral Supervisor at Jiangnan University, is a recipient of the Jiangsu Provincial Outstanding Youth Fund, a New Century Excellent Talent of the Ministry of Education, and a Level-2 Talent of Jiangsu Province's "333 High-Level Talent Training Program". He currently serves as Associate Director of the Jiangsu Key Laboratory of Human-Machine Fusion Software and Media Technology (a key laboratory of Jiangsu provincial universities). In recent years, his main research focuses on uncertain/explainable artificial intelligence and its applications in food bio-health computing, etc. He has published over 200 papers in relevant fields, including more than 70 regular papers in IEEE/ACM Transactions series. As a key contributor, he has won several scientific research awards such as the First Prize of Science and Technology Progress of the Ministry of Education, the First Prize of Natural Science of Zhejiang Province, and the Second Prize of Natural Science of the Ministry of Education. He has guided 8 graduate students to win Jiangsu Provincial Excellent Master's/Doctoral Theses. He has conducted research for three years at The Hong Kong Polytechnic University, University of California, Davis, and Osaka Prefecture University. He has served as a Member of the Hybrid Intelligence Professional Committee of the Chinese Association of Automation, a Member of the Granular Computing and Knowledge Discovery Professional Committee of the Chinese Association for Artificial Intelligence, a Member of the Fuzzy Logic and Many-Valued Logic Professional Committee of the China Computer Federation (CCF), an Executive Member of the Bioinformatics Professional Committee of CCF, an Executive Member of the Big Data Expert Committee of CCF, and Executive Director of the Youth Working Committee of Jiangsu Computer Society, among other positions.
Title: Explainable Artificial Intelligence: Machine Learning Driven by the Combination of Rules, Fuzzy Reasoning, and Data
Abstract: Explainable Artificial Intelligence (XAI) has emerged as a crucial development trend in the next generation of AI. Its research primarily focuses on two directions: 1) Post-hoc explainability techniques for black-box models such as deep learning; 2) Transparent intelligent model modeling techniques. From the perspective of uncertainty modeling based on rules and fuzzy reasoning, this report presents the speaker’s recent research findings in the second direction of explainable AI modeling, including rule-based and uncertainty reasoning-driven transfer learning, multi-view learning, representation learning, multi-label learning, graph data learning, as well as the applications of these methods in bioinformatics, biomedicine, and other fields.
Prof. Jia Wu
Monash University, Australia
Wu Jia, Ph.D. in Engineering, is a Professor at Monash University in Australia and a doctoral supervisor. As the Director of the Medical Information and Artificial Intelligence Data Decision-Making Laboratory in Australia, he serves as an IBM Medical Information Strategic Scientist, a Highly Cited Researcher by Clarivate, and a Highly Cited Scholar by Elsevier. He has received the Victoria Awards for Science and Technology from the Australian Academy of Science and has led projects such as the Australian Scholar Initiative Fund. His primary research focuses include medical information and artificial intelligence, healthcare big data decision-making, and medical imaging with artificial intelligence.
Title: Research on Artificial Intelligence Data Decision-Making in Medical Information-Assisted Diagnosis
Abstract: The interdisciplinary field of medical informatics and engineering is an internationally recognized applied frontier discipline. It integrates research methodologies such as medical informatics, medical imaging, artificial intelligence, machine learning, natural language processing, neural network optimization algorithms, medical and information decision-making, auxiliary diagnosis and fuzzy inference analysis. This discipline provides auxiliary diagnosis and rapid decision-making support for medical institutions and healthcare professionals, effectively reducing misdiagnosis rates in developing countries where doctors face large patient volumes. Moreover, it establishes medical imaging-based healthcare assistance to better aid doctors in analyzing and diagnosing patient conditions, thereby minimizing misdiagnosis rates and serving as an effective means for precision medicine and precise management. Through probabilistic analysis and decision-making models in artificial intelligence, it develops pathological staging prediction methods for auxiliary diagnosis, creating personalized medical databases tailored to individual patient conditions and characteristics. This enables adaptive treatment plan analysis and recommendation methods for clinicians. For clinical patients under physician management, interdisciplinary research combining medical imaging and related fields tracks and adjusts treatment plans in real-time for different pathological states during therapy cycles, establishing multi-source data collaboration for auxiliary treatment recommendation decision-making. This research direction focuses on future technological advancements, emphasizing team innovation capacity development and refining a series of new achievements.
Prof. Haijun Zhang
Harbin Institute of Technology, Shenzhen, China
Haijun Zhang received the B.Eng. and Master’s degrees from Northeastern University, Shenyang, China, and the Ph.D. degree from the Department of electronic Engineering, City University of Hong Kong, Hong Kong, in 2004, 2007, and 2010, respectively. He was a Post-Doctoral Research Fellow with the Department of Electrical and Computer Engineering, University of Windsor, Windsor, ON, Canada, from 2010 to 2011. Since 2012, he has been with the Shenzhen Graduate School, Harbin Institute of Technology, China, where he is currently a Professor of Computer Science. His current research interests include data mining, machine learning, generative AI, and embodied intelligence. He published over 180 technical papers in international journals and conferences. Prof. Zhang is currently a Senior Editor of IEEE Trans. on Consumer Electronics, etc.
Title: From Digitalization to Intelligentization: The AI-driven Technical Transformation of the Apparel Industry
Abstract: "Clothing," as one of the basic needs, plays an important role in human life. Following the emergence of the concept of the metaverse, generative large models such as GPT, as new tools marking the milestone of civilian AI, have attracted extensive attentions in both academia and industrial practitioners. This talk will cover the topics and basic tasks of fashion intelligence associated with its recent progresses and how AI is shaping the apparel industry. In particular, this talk showcases the trends and applications of fashion intelligence driven by multi-modal large language models, such as image and video generations. Moreover, the speaker will report some results from his research group, including controllable editing, collocation generation and recommendation, Mannequin2Real applications, cross-scene object layout, etc. At last, this talk will give some insights on the future prospect of fashion AI and the apparel industry.