Xiaoou Tang
Also known as Tang Xiaoou
Xiaoou Tang was a computer vision researcher whose work included face recognition and face attributes. He earned degrees at USTC, Rochester and MIT, then joined CUHK and established its Multimedia Laboratory. From 2005 to 2008, he led Microsoft Research Asia’s Visual Computing Group; in 2014, he co-founded SenseTime and his team reached 99.15% face-recognition accuracy. His team won the 2009 CVPR Best Paper Award, and he became an IEEE Fellow in 2009; he later co-authored a 2015 ICCV paper on face attributes and held leadership roles at CUHK, Pujiang Lab and Shanghai AI Lab. He died in 2023; the CelebA team received the PAMI Everingham Prize in 2024.
- Works
- 340
- Citations
- 99,379
- h-index
- 115
Publication numbers from OpenAlex, as of 14 October 2026.
Timeline
-
1990–1996
From USTC to MIT
He earned a BS at USTC in 1990, an MS at Rochester in 1991, and a PhD at MIT in 1996.
-
1998–2008
A lab and an industry role
He joined CUHK’s Information Engineering Department in 1998 and established its Multimedia Laboratory in 2001. From 2005 to 2008, he led the Visual Computing Group at Microsoft Research Asia.
-
2009
Awards in 2009
His team won the CVPR Best Paper Award. IEEE named him a Fellow in 2009.
-
2014
A face recognition record
His team developed a face recognition system that reached 99.15% accuracy. He co-founded SenseTime in 2014.
-
2015
A paper on face attributes
He co-authored the 2015 ICCV paper “Deep Learning Face Attributes in the Wild” with Ziwei Liu, Ping Luo and Xiaogang Wang.
-
By 2023
Leading AI laboratories
By 2023, he was a CUHK professor and department chair and directed Pujiang Lab and Shanghai AI Lab. He had also served on the HKSTP board.
-
2023
His death
He died in 2023.
-
2024
A prize for CelebA
At ECCV 2024, the PAMI Everingham Prize went to the CelebA Team, which listed Xiaoou Tang among its members.
Sources and credits
Photo: ie.cuhk.edu.hk; TANG Xiaoou, Sean - IE WEB; creator not stated. License: Not stated by source. Source: ie.cuhk.edu.hk. Cropped to a square and resized. Ask us to remove or replace this photo.