Zhixuan Li

I am a Research Fellow in the College of Computing and Data Science (CCDS) at Nanyang Technological University, collaborating with Prof. Weisi Lin. I received my Ph.D. degree at the National Engineering Research Center of Visual Technology (NERCVT), Peking University in 2023, supervised by Prof. Tiejun Huang and Assoc. Prof. Tingting Jiang. I received my B.E. degree from Tianjin University in 2018, supervised by Prof. Pengfei Zhu and Prof. Di Jin.

Google Scholar LinkedIn Email

Research map

From the known to the unknown.

My research develops AI systems that infer the unknown from the known—recovering hidden visual structures with vision-language models, imputing missing Earth observations, and forecasting future states from past dynamics.

Amodal Perception and Reasoning

Using vision-language models to infer complete object structures and reason about complex occlusions beyond visible evidence.

Amodal segmentation and reasoning under complex occlusion
VLM-guided shape completion from weak visual and language cues

Earth Observation

Reconstructing missing Earth observations under cloud occlusion and forecasting future environmental states from known spatial and temporal context.

Climate and Earth observation imputation under large-scale missingness
Weather-field and long-term time-series forecasting

Selected work

Publications

Trajectory

Experience

2024.03 — Present

Research Fellow

College of Computing and Data Science, Nanyang Technological University · Singapore

2018.09 — 2023.07

Doctor of Philosophy

School of Computer Science, Peking University · Beijing, China

2014.09 — 2018.07

Bachelor of Engineering

School of Computer Science, Tianjin University · Outstanding Graduates · Tianjin, China

Academic community

Service

Journal reviewer

IEEE TNNLS · IEEE TCSVT · ACM JATS · JVCI · IMAVIS

Conference reviewer