Jiajun Wu is a third-year Ph.D. student at Massachusetts Institute of Technology, advised by Professor Bill Freeman and Professor Josh Tenenbaum. His research interests lie on the intersection of computer vision, machine learning, and computational cognitive science. Before coming to MIT, he received his B.Eng. from Tsinghua University, China, advised by Professor Zhuowen Tu.

Research Summary

Humans demonstrate remarkable abilities in predicting physical events in dynamic scenes, and in inferring both geometric and physical object properties from visual input. There is also evidence that babies form a visual understanding of basic concepts of geometry and physics at an early age. These facts suggest the importance of building a machine with such competency.

Jiajun’s research focuses on visual understanding, especially on building models to characterize object properties from synthetic and real-world scenes. He has proposed datasets and methods for recognizing and synthesizing 3D shapes, and for perceiving physical object properties from static images or videos. He also studies learning from multi-modal data and incorporating learned knowledge into high-level reasoning among objects, scenes, and agents.

For more information, please visit his website: https://jiajunwu.com