Prof. Jianping Gou

IEEE Senior Member

Southwest University, China

Biography: 

Jianping Gou, Ph.D. in Engineering, Professor, Ph.D. Supervisor, has been selected as an Outstanding Young Core Instructor under the “Qing Lan Project” of Jiangsu Higher Education Institutions, a recipient of the International Exchange Program for Postdoctoral Researchers of China, and a talent under the Chongqing Overseas Returnee Entrepreneurship and Innovation Support Program. He serves as an Executive Committee Member of the Technical Committee on Multimedia, the Technical Committee on Computer Vision, and the Youth Working Committee of the Artificial Intelligence Society. He is also an IEEE Senior Member, a Senior Member of the China Society of Image and Graphics (CSIG), and a Senior Member of the China Computer Federation (CCF).

Professor Gou’s research focuses on theories and applications in machine learning and pattern recognition. He has published more than 140 academic papers in well-known journals and conferences, including IJCV and ACM/IEEE Transactions, among which over 90 are indexed in SCI. His publications include 4 highly cited papers and 1 hot paper, with a single paper receiving nearly 2,000 citations on Google Scholar. He has led 6 research projects, including general and youth programs of the National Natural Science Foundation of China as well as provincial and ministerial-level projects. He holds 8 authorized national invention patents. His honors include the First Prize for Excellent Paper at the Sichuan-Chongqing Science and Technology Academic Conference, the Third Prize of Jiangsu Science and Technology Award, and the Third Prize of Hospital Science and Technology Innovation Award of the Chinese Hospital Association.

Speech Title:

Diversity-driven Knowledge Distillation for Large-scale Model Compression

Abstract:
Knowledge distillation for model compression is a core technology of empowering large-scale models for various downstream applications with low-cost and high-efficiency. On the basis of briefly introducing the development of large-scale models and summarizing the relevant technologies of large model compression, the theory, algorithms, and applications of model distillation are reviewed, the series of works on diversity-driven knowledge distillation are further presented, and the latest large language model distillation is reported. Finally, the prospects of large-scale model distillation are given.

Prof. Yue Liu

Beijing Institute of Technology, China

Biography: 

Yue Liu is a Professor and Ph.D. supervisor at the School of Optics and Photonics, Beijing Institute of Technology. He serves as Director of the Institute of Optoelectronic Information Technology and Color Engineering and Deputy Director of the Beijing Engineering Research Center of Mixed Reality and Novel Display. He is also a member of the expert panel for the National Key R&D Program on “Culture, Science & Technology and Modern Service Industries.” In 2024, he was selected for a national-level innovation talent support program.

His main research areas include virtual reality and augmented reality, natural human–computer interaction, and computer vision. He currently serves as Vice Chair of the Technical Committee on Intelligent Interaction of the Chinese Association for Artificial Intelligence; Vice Chair of the Technical Committee on Human–Computer Interaction of the China Computer Federation; Vice Chair of the Virtual Reality Industry Branch of the Chinese Institute of Electronics; and Standing Board Member, Deputy Secretary-General, and Chair of the Virtual Reality Technical Committee of the China Society of Image and Graphics. He has previously served as Vice Chair of the Virtual Technology and Applications Technical Committee and Vice Chair of the Virtual Simulation Education and Teaching Technical Committee of the Chinese Association for Simulation, as well as Vice Chair of the Virtual Reality Technical Committee of the China Computer Federation.

He is currently the Chief Scientist of a National Key R&D Program project under the 14th Five-Year Plan. He has published more than 100 papers and filed over 50 patents. His research has been recognized with a Second Prize of the National Award for Technological Invention.

Speech Title:

Multi-Scale Intrinsic Image Reconstruction for Applications Driven by Intrinsic Imaging Properties

Abstract:
As an important intermediate representation connecting observed images with real scene properties, intrinsic images are valuable for applications such as image editing, augmented reality, digital restoration, and material inspection. To meet the needs of both local fine-scale imaging and global complex scene modeling, efficient and stable intrinsic image reconstruction is required across multiple scales. This report focuses on application demands driven by intrinsic imaging properties and introduces our research and practice in multi-scale intrinsic image reconstruction.

Assoc. Prof. Tao Zhang

Shanghai Jiao Tong University, China

Biography: 

Tao Zhang is an Associate Professor at the School of Automation and Intelligent Sensing, Shanghai Jiao Tong University, who is also an Associate Editor of IEEE Transactions on Image Processing and the Secretary-General of the Shanghai Chapter of IEEE GRSS. His research focuses on pattern recognition, SAR image scene interpretation, and land cover/landform change detection. He has led projects including the National Natural Science Foundation of China (NSFC) Young Scientist Fund, China Postdoctoral Science Foundation, and National Key Laboratory Open Fund, and participated in multiple research initiatives including NSFC Major/Key/General Projects. He has published over 70 SCI-indexed papers in renowned domestic and international journals (including 4 ESI highly cited papers) and holds 5 national invention patents (applied/authorized), which contribute to many awards, including the First Prize for Scientific and Technological Progress from the Shanghai Automation Society, the First Prize for Service Industry Technological Innovation from the China General Chamber of Commerce, the Second Prize for Teaching Achievement from the Shanghai Computer Society and so on.

Speech Title:

Polarization Information-Driven Ship Target Detection Methods in SAR Images

Abstract:
Synthetic aperture radar (SAR) is a microwave imaging system capable of all-day Earth observation. In recent years, ship target detection using SAR images has attracted extensive attention. High-precision ship detection can effectively support maritime traffic control, illegal fishing supervision, maritime search and rescue, and other related applications. Compared with single-polarization SAR data, multi-polarization SAR, especially fully polarimetric SAR (PolSAR), can simultaneously capture the backscattering intensity and phase information of targets, enabling better characterization of backscattering differences between targets and clutter. To improve the detection accuracy of ship targets in SAR imagery, we focus on the utilization and mining of polarimetric information and have proposed a variety of polarimetric-information-based backscattering characteristic description matrices, such as the [P] matrix and [CP] matrix. Experiments on multiple diverse datasets have validated the effectiveness of the proposed methods. Notably, for weak ship targets, the Target-to-Clutter Ratio (TCR) can be significantly enhanced.