Miao LIU
Assistant Professor
Engaged in the research of first-person vision, augmented reality, and generative AI models.
Education/Work Experience
June 2015: Bachelor's degree from Beihang University (Beijing University of Aeronautics and Astronautics).
May 2017: Master's degree from Carnegie Mellon University, USA.
June 2022: Ph.D. from the Georgia Institute of Technology, USA.
August 2022-June 2024: Research Scientist at Meta, GenAI.
June 2024-Present: Senior Research Scientist at Meta, GenAI.
Research Achievements
As the primary author, participated in the construction of several influential first-person video datasets, including Eg04D, EgOEx04D, EGTEAGaze+, and Behavior Vision Suite. These achievements have been widely applied in both academia and industry.
Proposed and designed multiple algorithms for behavior understanding and prediction in first-person videos, some of which have been successfully deployed in Meta Reality Labs' next-generation smart glasses products.
During the Meta GenAI period, participated in the end-to-end development of generative AI models such as EMU, Llama3, and Llama4 (only the multimodal part).
Published more than 20 papers as the primary author in top conferences/journals such as CVPR, ECCV, ACL, and TPAMI. The research findings have been cited over 7,800 times, with eight papers selected for oral presentations. The CVPR 2022 and ECCV 2024 papers were shortlisted for best paper candidates, and the BMVC 2022 paper received the Best Student Paper Award. The academic achievements and development potential are outstanding.