Shiming Ge | AI Safety | Best Researcher Award

Prof. Shiming Ge | AI Safety | Best Researcher Award

Professor at Chinese Academy of Sciences, China

Dr. Shiming Ge is a prominent researcher and academic leader in the field of artificial intelligence, with specializations in AI safety, multimedia security, visual understanding, and privacy-preserving machine learning. He serves as a Professor and the Head of the Intelligent Multimedia Security Lab at the Institute of Information Engineering, Chinese Academy of Sciences (CAS). Dr. Ge has made significant contributions to areas such as deepfake detection, robust learning from noisy data, and adversarial machine learning. His work bridges the gap between theoretical AI foundations and real-world applications, addressing modern technological challenges. With over 100 research papers in top-tier journals and conferences, he is recognized both nationally and internationally. He has also successfully led numerous national and industry-funded research projects. Dr. Ge is a Senior Member of IEEE and plays an active role in mentoring young researchers, shaping the future of AI research through leadership, innovation, and global academic engagement.

Professional Profile 

Education🎓

Dr. Shiming Ge pursued his higher education through an accelerated track, showcasing exceptional academic performance from an early stage. He earned his Ph.D. in Electronic Engineering from the University of Science and Technology of China (USTC) in 2008. Before his doctoral studies, he completed his bachelor’s degree in Electronic Engineering from USTC as part of an elite early-excellence program that allowed him to graduate one year ahead of schedule. His Ph.D. research focused on digital image recovery, laying a solid technical foundation that would later underpin his work in computer vision, machine learning, and multimedia analysis. His advanced education provided him with strong theoretical and practical expertise, which he has successfully applied in tackling pressing issues such as visual security, data privacy, and adversarial robustness. Dr. Ge’s academic journey reflects not only his intellectual capacity but also his long-term commitment to innovation and academic excellence in artificial intelligence.

Professional Experience📝

Dr. Shiming Ge currently holds the position of Professor and leads the Intelligent Multimedia Security Lab at the Institute of Information Engineering, Chinese Academy of Sciences (CAS). Over the years, he has taken on several prominent roles within the academic and industrial research community, contributing to both theoretical advancements and real-world applications. He has served as the principal investigator for more than 20 national and industrial research projects, including those funded by China’s National Key R&D Program and companies such as Siemens and Alibaba. His leadership in research and project management has yielded numerous successful outcomes in AI and multimedia security. Beyond project execution, he also teaches graduate-level courses, such as “AI Safety” and “Introduction to Deep Learning,” and actively mentors young researchers. His students have gone on to win prestigious awards such as the CAS President’s Award. His professional journey highlights a balanced dedication to academic innovation, industry collaboration, and talent development.

Research Interest🔎

Dr. Shiming Ge’s research interests span several high-impact domains within artificial intelligence, with a central focus on AI safety, multimedia security, visual understanding, adversarial machine learning, and privacy-preserving techniques. He is particularly known for his work on deepfake detection, robust learning under noisy conditions, and trustworthy machine learning—topics that are increasingly critical in today’s digital age. He has also contributed to the development of federated learning frameworks and few-shot learning models, enabling decentralized and efficient AI applications. His research addresses both theoretical challenges and practical implementations, emphasizing the societal importance of trustworthy and secure AI systems. Dr. Ge is also exploring the interpretability of neural networks to improve transparency and accountability in automated decision-making. His work has not only advanced the academic field but also had tangible impacts on real-world technologies, such as surveillance, identity verification, and online content moderation, making his research highly relevant and applicable.

Award and Honor🏆

Dr. Shiming Ge has received multiple prestigious awards in recognition of his groundbreaking research in artificial intelligence and multimedia security. He was honored with the IEEE TMM Best Journal Paper Award in 2025, underscoring the quality and impact of his published work. In addition, he is a recipient of the Wu Wenjun Artificial Intelligence Science and Technology Award, one of China’s most distinguished recognitions in the field of AI. His research team has earned top rankings in international competitions, including the CVPR’20 Anti-UAV Challenge and the ICCV’19 VisDrone-SOT Challenge, reflecting his capacity to translate research into high-performing systems. Dr. Ge’s students have also gained accolades under his mentorship, such as the CAS President’s Award and recognition as Outstanding Graduates of Beijing, further highlighting his excellence in academic supervision. These awards collectively affirm his leadership, innovation, and sustained contributions to advancing AI technologies globally.

Research Skill🔬

Dr. Shiming Ge possesses a comprehensive and multidisciplinary research skillset that enables him to lead innovative projects across multiple AI domains. He is proficient in deep learning frameworks such as TensorFlow and PyTorch, and has extensive experience in designing robust neural network architectures for tasks like object detection, face recognition, and anomaly detection. His technical expertise extends to adversarial training, federated learning, and privacy-preserving machine learning. Dr. Ge also excels in multimedia data analysis, image forensics, and deepfake identification techniques. Beyond technical implementation, he is highly skilled in hypothesis-driven research design, scientific writing, and cross-functional project coordination. His strong grasp of statistical modeling, algorithm optimization, and system-level integration has enabled the successful deployment of AI models in both academic and industry contexts. Furthermore, he has demonstrated a capacity for mentoring, peer reviewing, and leading multidisciplinary teams, making him a valuable contributor to collaborative and large-scale research environments.

Conclusion💡

Publications Top Noted

  • Title: Detecting masked faces in the wild with LLE-CNNs
    Authors: S. Ge, J. Li, Q. Ye, Z. Luo
    Year: 2017
    Citations: 576

  • Title: Selective-supervised contrastive learning with noisy labels
    Authors: S. Li, X. Xia, S. Ge, T. Liu
    Year: 2022
    Citations: 300

  • Title: Low-resolution face recognition in the wild via selective knowledge distillation
    Authors: S. Ge, S. Zhao, C. Li, J. Li
    Year: 2018
    Citations: 236

  • Title: Occluded face recognition in the wild by identity-diversity inpainting
    Authors: S. Ge, C. Li, S. Zhao, D. Zeng
    Year: 2020
    Citations: 130

  • Title: Student network learning via evolutionary knowledge distillation
    Authors: K. Zhang, C. Zhang, S. Li, D. Zeng, S. Ge
    Year: 2021
    Citations: 105

  • Title: Detecting Deepfake Videos with Temporal Dropout 3DCNN
    Authors: D. Zhang, C. Li, F. Lin, D. Zeng, S. Ge
    Year: 2021
    Citations: 99

  • Title: Estimating noise transition matrix with label correlations for noisy multi-label learning
    Authors: S. Li, X. Xia, H. Zhang, Y. Zhan, S. Ge, T. Liu
    Year: 2022
    Citations: 98

  • Title: Accurate temporal action proposal generation with relation-aware pyramid network
    Authors: J. Gao, Z. Shi, G. Wang, J. Li, Y. Yuan, S. Ge, X. Zhou
    Year: 2020
    Citations: 97

  • Title: Domain adaptive attention learning for unsupervised person re-identification
    Authors: Y. Huang, P. Peng, Y. Jin, Y. Li, J. Xing
    Year: 2020
    Citations: 89

  • Title: Look through masks: Towards masked face recognition with de-occlusion distillation
    Authors: C. Li, S. Ge, D. Zhang, J. Li
    Year: 2020
    Citations: 87

  • Title: Efficient low-resolution face recognition via bridge distillation
    Authors: S. Ge, S. Zhao, C. Li, Y. Zhang, J. Li
    Year: 2020
    Citations: 82

  • Title: Predicting aesthetic score distribution through cumulative Jensen-Shannon divergence
    Authors: X. Jin, L. Wu, X. Li, S. Chen, S. Peng, J. Chi, S. Ge, C. Song, G. Zhao
    Year: 2018
    Citations: 82

  • Title: Bootstrapping multi-view representations for fake news detection
    Authors: Q. Ying, X. Hu, Y. Zhou, Z. Qian, D. Zeng, S. Ge
    Year: 2023
    Citations: 80

  • Title: VisDrone-SOT2019: The vision meets drone single object tracking challenge results
    Authors: D. Du, P. Zhu, L. Wen, X. Bian, H. Ling, Q. Hu, J. Zheng, T. Peng, X. Wang, …
    Year: 2019
    Citations: 75

  • Title: ILGNet: Inception modules with connected local and global features for efficient image aesthetic quality classification using domain adaptation
    Authors: X. Jin, L. Wu, X. Li, X. Zhang, J. Chi, S. Peng, S. Ge, G. Zhao, S. Li
    Year: 2019
    Citations: 73

  • Title: Attentive Deep Stitching and Quality Assessment for 360 Omnidirectional Images
    Authors: J. Li, Y. Zhao, W. Ye, K. Yu, S. Ge
    Year: 2019
    Citations: 63

  • Title: Aesthetic attributes assessment of images
    Authors: X. Jin, L. Wu, G. Zhao, X. Li, X. Zhang, S. Ge, D. Zou, B. Zhou, X. Zhou
    Year: 2019
    Citations: 59

  • Title: The 3rd anti-UAV workshop & challenge: Methods and results
    Authors: J. Zhao, J. Li, L. Jin, J. Chu, Z. Zhang, J. Wang, J. Xia, K. Wang, Y. Liu, …
    Year: 2023
    Citations: 53

  • Title: Deepfake video detection with spatiotemporal dropout transformer
    Authors: D. Zhang, F. Lin, Y. Hua, P. Wang, D. Zeng, S. Ge
    Year: 2022
    Citations: 51

  • Title: Deepfake video detection via predictive representation learning
    Authors: S. Ge, F. Lin, C. Li, D. Zhang, W. Wang, D. Zeng
    Year: 2022
    Citations: 50