Jingcheng Ke | Diffusion Models | Excellence in Research

Dr. Jingcheng Ke, Osaka university, Japan.

Jingcheng Ke, Ph.D. , is a researcher at the Institute for Datability Science, Osaka University . With a Ph.D. from National Tsing Hua University (NTHU) , his research focuses on vision-language matching and diffusion models for image and video analysis . He has worked as an AI researcher at vivo AI Lab and as an exchange student at Shenzhen Key Laboratory of Visual Object Detection and Recognition. Proficient in multiple languages and programming , Dr. Ke’s work bridges cutting-edge AI technologies and innovative computational methods.

Publication Profile

Googlescholar

Education & Experience:

Education

Ph.D. in Communications Engineering (2019–2024)
- National Tsing Hua University, Taiwan
- Thesis: Referring Expression Comprehension in a Graph-based Perspective and Its Generalizations
M.Sc. in Computer Application (2015–2018)
- Shaanxi Normal University, China
- Thesis: Face recognition based on virtual faces and sparse representations
B.Sc. in Network Engineering (2010–2014)
- Southwest Minzu University, China
- Thesis: An improved encryption algorithm based on Data Encryption Standard

Experience

Researcher (2024–Present)
- Institute for Datability Science, Osaka University
AI Researcher (2018–2019)
- vivo AI Lab
Exchange Student (2016–2018)
- Shenzhen Key Laboratory of Visual Object Detection and Recognition

Suitability for the Award

Dr. Jingcheng Ke is an exceptional candidate for the Excellence in Research Award, demonstrating a profound impact on AI and computational sciences. His Ph.D. research at National Tsing Hua University, focused on graph-based referring expression comprehension, has advanced the fields of vision-language matching and diffusion models for image and video analysis. With professional experience at Osaka University and vivo AI Lab, Dr. Ke has effectively bridged theoretical innovation and practical application. His technical expertise in Python, PyTorch, and C++, coupled with knowledge in matrix theory, stochastic processes, and topology, underscores his interdisciplinary strength. Dr. Ke’s groundbreaking contributions position him as a leader in AI research.

Professional Development

Dr. Jingcheng Ke’s professional journey spans academia and industry, specializing in artificial intelligence and computer vision . His Ph.D. research at NTHU explored graph-based perspectives for referring expression comprehension, advancing the intersection of vision and language technologies . With hands-on experience in AI innovation at vivo AI Lab and collaboration with top-tier research labs, he has honed his expertise in diffusion models and image/video analysis . Proficient in coding languages like Python and PyTorch , he leverages advanced mathematical concepts like matrix theory and stochastic processes to push AI boundaries .

Research Focus

Dr. Ke’s research is centered on the intersection of vision and language , with a keen focus on diffusion models for image and video analysis . His work addresses challenges in vision-language matching, exploring graph-based approaches to enhance comprehension and generalization capabilities . Passionate about advancing AI technologies, he delves into areas like sparse representation and encryption algorithms . By integrating robust coding skills in Python and PyTorch with theoretical foundations, his research contributes to groundbreaking advancements in artificial intelligence and computational methodologies .

Awards and Honors

Best Paper Award – Recognized for excellence in vision-language research.
Graduate Fellowship – National Tsing Hua University, Taiwan.
Outstanding Thesis Award – Shaanxi Normal University, China.
Research Excellence Recognition – vivo AI Lab, 2019.
Academic Merit Scholarship – Southwest Minzu University, China.

Publication Highlights

An improvement to linear regression classification for face recognition – 26 citations, published in International Journal of Machine Learning and Cybernetics, 2019.
Referring Expression Comprehension via Enhanced Cross-modal Graph Attention Networks – 12 citations, published in ACM TOMM, 2022.
Face recognition based on symmetrical virtual image and original training image – 12 citations, published in Journal of Modern Optics, 2018.
Sample partition and grouped sparse representation – 8 citations, published in Journal of Modern Optics, 2017.
A novel grouped sparse representation for face recognition – 7 citations, published in Multimedia Tools and Applications, 2019.