| CARVIEW |
Select Language
HTTP/2 200
server: GitHub.com
content-type: text/html; charset=utf-8
last-modified: Thu, 09 Mar 2023 07:43:20 GMT
access-control-allow-origin: *
etag: W/"64098e18-6ae3"
expires: Mon, 29 Dec 2025 22:42:14 GMT
cache-control: max-age=600
content-encoding: gzip
x-proxy-cache: MISS
x-github-request-id: E1AB:3FD64F:962321:A876EB:6953016D
accept-ranges: bytes
age: 0
date: Mon, 29 Dec 2025 22:32:14 GMT
via: 1.1 varnish
x-served-by: cache-bom-vanm7210042-BOM
x-cache: MISS
x-cache-hits: 0
x-timer: S1767047534.387599,VS0,VE206
vary: Accept-Encoding
x-fastly-request-id: 0dcf43fb6f2d6becc8fe2637e1674067a416b978
content-length: 7852
index
Yichen Wei (危夷晨)
切换至中文主页。
Vice President of Research and Development of SHUKUN Technology
Email: weiyc at shukun.net
I joined SHUKUN Technology on November, 2021. Before that, I was the director of Megvii (Face++) Research Shanghai for 3 years and a researcher in Visual Computing group, Microsoft Research Asia (MSRA) for 12 years. I received my Ph.D degree in Hong Kong University of Science and Technology in 2006, and B.S. degree in Peking University in 2001, respectively.
My research interests include deep learning and computer vision.
My Google Scholar citation is about 23,000, h-index is 48.
My work in MSRA has been transferred to Kinect Identity in XBox, Windows Hello, Microsoft Cognitive Service, Bing, Office, and Microsoft XiaoIce, etc.
News
- March 2022, I gave a talk on "Introduction of Application of AI in Medical Images" in 机器之心AI科技年会.
- November 2019, I gave a tech talk on "uncertainty learning in computer vision" in 将门创投.
- April 2019, my work on AutoML is reported (新智元, 机器之心, 量子位, ...).
- Jan 2019, I was interviewed by "数据科学50人" program.
- July 2018, I was interviewed by 新智元.
Research Highlights
- Deformable Convolutional Networks ranks 6th of the most influential papers in ICCV 2017 and won 3rd place in object detection task of COCO Challenge 2017.
- Single path one-shot neural architecture search with uniform sampling ranks 9th of the most influential papers in ECCV 2020.
- Simple Baselines For Human Pose Estimation And Tracking ranks 12th of the most influential papers in ECCV 2018, won 1st place in the PoseTrack2018 Multi-person Pose Tracking Challenge and 2nd place of keypoint detection task of COCO 2018.
- Face Alignment by Explicit Shape Regression ranks 12th of the most influential papers in CVPR 2012.
- Saliency Optimization from Robust Background Detection ranks 14th of the most influential papers in CVPR 2014.
- Rethinking on Multi-Stage Networks for Human Pose Estimation won 1st place in keypoint detection task of COCO and Mapillary Challenge 2019.
- Fully Convolutional Instance-aware Semantic Segmentation won 1st place in object segmentation task of COCO Challenge 2016.
- Integral Human Pose Regression won the 2nd place of ECCV 2018 3D Human Pose Estimation Challenge.
Publications in chronology
Publications in topic
Last Modification: March 9th, 2023