Jingwei Ji

Jingwei Ji is a Research Scientist at Waymo LLC, specializing in computer vision and machine learning. He holds a Ph.D. in Electrical Engineering from Stanford University, where he was co-advised by Prof. Juan Carlos Niebles and Prof. Silvio Savarese, and a B.Sc. in Physics from Peking University.

• Education
  • Ph.D. in Electrical Engineering, Stanford University
  • B.Sc. in Physics, Peking University

His research focuses on human activity understanding, video analysis, 3D vision, and scene graph modeling. He pioneered the Action Genome dataset, which bridges human actions and human-object relationships through spatio-temporal scene graphs.

Jingwei's recent publications emphasize temporal alignment in few-shot learning, semantic segmentation of actors/actions, and 3D shape reconstruction. His work spans both theoretical advancements and practical applications in autonomous systems.

Academic services include serving as a reviewer for top-tier conferences like CVPR, ICCV, AAAI, and journals including The Visual Computer. He organized workshops at CVPR 2020 and ECCV 2020 on compositional perception.

Previously, he worked as an Applied Scientist Intern at Amazon Go (2018) and contributed to open-source projects like ActionGenome, a Python-based video dataset framework under MIT license.