危夷晨

旷视科技上海研究院院长

电子邮件: weiyichen at megvii.com

我于2018年7月加入旷视科技，任上海研究院院长。在此之前, 我在微软亚洲研究院视觉计算组工作了12年。我于2006年在香港科技大学获得博士学位，于2001年在北京大学获得学士学位。

我的研究方向为深度学习和计算机视觉。

我的谷歌学术引用数大概为10,000, h指数为38。

我在微软期间的研究工作成功转化到微软的产品，包括Kinect 身份识别, Windows Hello, 微软认知服务, 必应, Office, and 微软小冰, 等等。

上海研究院的工作岗位长期开放（研究员/工程师/实习）

有意向者请发简历给我，主题为“旷视上海研究院工作申请”或者“旷视上海研究院实习申请”。

动态

• 2020年3月, 旷视的深度学习框架MegEngine (天元) 在GitHub开源。
• 2019年11月，我在将门创投进行了一个技术宣讲 ，主题为 "不确定性估计在计算机视觉中的应用"。
• 2019年4月，我们关于AutoML的工作获得一些公众报道（新智元, 机器之心, 量子位, ...）。
• 2019年1月, 我接受了"数据科学50人" 栏目的采访。
• 2019年7月，我接受了新智元的采访。

计算机视觉竞赛获奖

• COCO and Mapillary Challenge 2019, 1st place in keypoint detection task with Rethinking on Multi-Stage Networks for Human Pose Estimation
• Our integral pose regression method has won the 2nd place of ECCV 2018 3D Human Pose Estimation Challenge
• Our simple baseline method has won the winner of PoseTrack2018 Multi-person Pose Tracking Challenge and the 2nd place of keypoint detection task of COCO 2018
• COCO Challenge 2017, 3rd place in object detection task with Deformable Convolutional Networks.
• COCO Challenge 2016, 1st place in object segmentation task with Fully Convolutional Instance-aware Semantic Segmentation.

最近的科研工作

Circle Loss: A Unified Perspective of Pair Similarity Optimization, CVPR 2020 (Oral)
Yifan Sun, Changmao Cheng, Yuhan Zhang, Chi Zhang, Liang Zheng, Zhongdao Wang, Yichen Wei
arXiv version

Data Uncertainty Learning in Face Recognition, CVPR 2020
Jie Chang, Zhonghao Lan, Changmao Cheng, Yichen Wei
arXiv version

Balanced Alignment for Face Recognition: A Joint Learning Approach
Huawei Wei, Peng Lu, Yichen Wei
Tech report on arXiv, March 2020

Towards Stabilizing Batch Statistics in Backward Propagation of Batch Normalization, ICLR 2020
Junjie Yan, Ruosi Wan, Xiangyu Zhang, Wei Zhang, Yichen Wei, Jian Sun
arXiv version

3D Dense Face Alignment via Graph Convolution Networks
Huawei Wei, Shuang Liang, Yichen Wei
Tech report on arXiv, April 2019

Single Path One-Shot Neural Architecture Search with Uniform Sampling
Zichao Guo, Xiangyu Zhang, Haoyuan Mu, Wen Heng, Zechun Liu, Yichen Wei, Jian sun
Tech report on arXiv, Mar 2019

Vehicle Re-Identification With Viewpoint-Aware Metric Learning, ICCV 2019
Ruihang Chu, Yifan Sun, Yadong Li, Zheng Liu, Chi Zhang, Yichen Wei
arXiv version

Re-Identification Supervised Texture Generation, CVPR 2019
Jian Wang, Yunshan Zhong, Yachun Li, Chi Zhang, Yichen Wei
arXiv version

Rethinking on Multi-Stage Networks for Human Pose Estimation
Wenbo Li, Zhicheng Wang, Binyi Yin, Qixiang Peng, Yuming Du, Tianzi Xiao, Gang Yu, Hongtao Lu, Yichen Wei, Jian Sun
Tech report on arXiv, Jan 2019, 1st winner for COCO 2019 keypoint detection

Towards High Performance Video Object Detection for Mobiles
Xizhou Zhu, Jifeng Dai, Xingchi Zhu, Yichen Wei, Lu Yuan
Tech report on arXiv, April 2018

Simple Baselines for Human Pose Estimation and Tracking, ECCV 2018
Bin Xiao, Haiping Wu, Yichen Wei
arXiv version
Code

Learning Region Features for Object Detection, ECCV 2018
Jiayuan Gu, Han Hu, Liwei Wang, Yichen Wei, Jifeng Dai
arXiv version

Integral Human Pose Regression, ECCV 2018
Xiao Sun, Bin Xiao, Shuang Liang, Yichen Wei
arXiv version
Slides Code

Relation Networks for Object Detection, CVPR 2018(Oral)
Han Hu, Jiayuan Gu, Zheng Zhang, Jifeng Dai, Yichen Wei
arXiv version
Code

Towards High Performance Video Object Detection, CVPR 2018 (Spotlight)
Xizhou Zhu, Jifeng Dai, Lu Yuan, Yichen Wei
arXiv version

Pseudo-Mask Augmented Object Detection, CVPR 2018
Xiangyun Zhao, Shuang Liang, Yichen Wei
arXiv version

Deformable Convolutional Networks, ICCV 2017 (Oral) Jifeng Dai, Haozhi Qi, Yuwen Xiong, Yi Li, Guodong Zhang, Han Hu, and Yichen Wei
arXiv version
Code Slides COCO 2017 workshop A third party demon on MNIST

Flow-Guided Feature Aggregation for Video Object Detection, ICCV 2017
Xizhou Zhu, Yujie Wang, Jifeng Dai, Lu Yuan, Yichen Wei
arXiv version
Code Video

Compositional Human Pose Regression, ICCV 2017
Xiao Sun, Jiaxiang Shang, Shuang Liang, Yichen Wei
arXiv version
Slides Video

Towards 3D Human Pose Estimation in the Wild: A Weakly-Supervised Approach, ICCV 2017
Xingyi Zhou, Qixing Huang, Xiao Sun, Xiangyang Xue, Yichen Wei
arXiv version
Code

Deep Feature Flow for Video Recognition, CVPR 2017
Xizhou Zhu, Yuwen Xiong, Jifeng Dai, Lu Yuan, Yichen Wei
arXiv version
Code Video

Fully Convolutional Instance-aware Semantic Segmentation, CVPR 2017 (Spotlight)
Yi Li, Haozhi Qi, Jifeng Dai, Xiangyang Ji, Yichen Wei
arXiv version
Code

早期的科研工作

最近修改于: 2020年4月13日