Chunyu Wang (王春雨)

Senior Researcher

Microsoft Research Asia

Biography

I am a Senior Researcher in Intelligent Multimedia Group, Microsoft Research Asia. I work on computer vision problems with special focus on 2D and 3D human pose estimation and tracking. I obtained PhD degree from Peking University in 2016. My advisor is Prof. Yizhou Wang. I spent two years (2012 and 2014) at UCLA, working with Prof. Alan L. Yuille. I obtained my Bachelor degree from Software School of Dalian University of Technology.

NEWS

[2020-04-06] The code for “A Simple Baseline for Multi-Object Tracking” is released! (code pdf)
[2020-03-26] The code for “Fusing Wearable IMUs with Multi-View Images for Human Pose Estimation: A Geometric Approach, CVPR 2020” is released! (code)
[2019-10-01] The code for “Cross View Fusion for 3D Human Pose Estimation, ICCV 2019” is released! (code)
[2019-10-01] The code for “Optimizing Network Structure for 3D Human Pose Estimation, ICCV 2019” is released! (code)
[2019-10-01] The code "Semantic Image Segmentation by Scale-Adaptive Networks, TIP 2019" is released! (code)

DEMOS

Publications

A Simple Baseline for Multi-Object Tracking

This is a real-time multi-object tracker that ranks first in many MOT evaluation datasets.

Yifu Zhang, Chunyu Wang, Xinggang Wang, Wenjun Zeng, Wenyu Liu, (arxiv 2020)

PDF Code Poster Video

End-to-End Estimation of Multi-Person 3D Poses from Multiple Cameras

We present an approach to estimate 3D poses of multiple people from multiple camera views.

Hanyue Tu, Chunyu Wang, Wenjun Zeng, (arxiv 2020)

PDF Video

Fusing Wearable IMUs with Multi-View Images for Human Pose Estimation: A Geometric Approach

We present a novel way of using IMUs for human pose estimation.

Zhe Zhang, Chunyu Wang, Wenhu Qin, Wenjun Zeng, (CVPR 2020)

PDF Code Poster Video

MetaFuse: A Pre-trained Fusion Model for Human Pose Estimation

We introduce MetaFuse, a pre-trained fusion model learned from a large number of cameras in the Panoptic dataset.

Rongchang Xie, Chunyu Wang, Yizhou Wang, (CVPR 2020)

PDF Poster

Semantic Image Segmentation by Scale-Adaptive Networks

Semantic image segmentation is an important yet unsolved problem. One of the major challenges is the large variability of the object …

Zilong Huang, Chunyu Wang, Xinggang Wang, Wenyu Liu, Jingdong Wang, (TIP 2019)

PDF Code

Cross View Fusion for 3D Human Pose Estimation

We address the problem of recovering absolute 3D human poses from multi-view images by incorporating multi-view geometric priors into …

Haibo Qiu, Chunyu Wang, Jingdong Wang, Naiyan Wang, Wenjun Zeng, (ICCV 2019)

PDF Code Poster Video

Optimizing Network Structure for 3D Human Pose Estimation

We propose Locally Connected Network for 3D human pose estimation. It extracts and propagates features over the pose graph and can …

Hai Ci, Chunyu Wang, Xiaoxuan Ma, Yizhou Wang, (ICCV 2019)

PDF Code Poster Video

Object detection in videos by high quality object linking

we focus on obtaining high quality object linking results for better classification. Unlike previous methods that link objects by …

Peng Tang, Chunyu Wang, Xinggang Wang, Wenyu Liu, Wenjun Zeng, Jingdong Wang, (TPAMI 2019)

PDF

Learning Basis Representation to Refine 3D Human Pose Estimations

Estimating 3D human poses from 2D joint positions is an ill-posed problem, and is further complicated by the fact that the estimated 2D …

Chunyu Wang, Haibo Qiu, Alan L. Yuille, Wenjun Zeng, (AAAI 2019)

PDF Poster

Robust 3d human pose estimation from single images or video sequences

We propose a method for estimating 3D human poses from single images or video sequences.

Chunyu Wang, Yizhou Wang, Zhouchen Lin, Alan L. Yuille, (TPAMI 2018)

PDF

Online Dictionary Learning for Approximate Archetypal Analysis

We propose a variant of archetypal analysis which scales gracefully to large datasets. The core idea is to decouple the binding between …

Jieru Mei, Chunyu Wang, Wenjun Zeng, (ECCV 2018)

PDF Code Poster

Video object segmentation by learning location-sensitive embeddings

We address the problem of video object segmentation which outputs the masks of a target object throughout a video given only a bounding …

Hai Ci, Chunyu Wang, Yizhou Wang, (ECCV 2018)

PDF Code Poster Video Custom Link

Learning discriminative activated simplices for action recognition

We address the task of action recognition from a sequence of 3D human poses. This is a challenging task firstly because the poses of …

Chenxu Luo, Chang Ma, Chunyu Wang, Yizhou Wang, (AAAI 2017)

PDF

Mining 3D Key-Pose-Motifs for Action Recognition

A key-pose-motif contains a set of ordered poses or action units(a short sequence of poses), which are required to be close but not …

Chunyu Wang, Yizhou Wang, Alan L. Yuille, (CVPR 2016)

PDF

Recognizing actions in 3d using action-snippets and activated simplices

Pose-based action recognition in 3D is the task of recognizing an action (e.g., walking or running) from a sequence of 3D skeletal …

Chunyu Wang, John Flynn, Yizhou Wang, Alan L. Yuille, (AAAI 2016)

PDF

Robust Estimation of 3D Human Poses from a Single Image

We propose a method of estimating 3D human poses from a single image, which works in conjunction with an existing 2D pose detector.

Chunyu Wang, Yizhou Wang, Zhouchen Lin, Alan L. Yuille, Wen Gao, (CVPR 2014)

PDF Code

An approach to pose-based action recognition

We propose one of the earliest approach for pose-based action recognition in wild videos.

Chunyu Wang, Yizhou Wang, Alan L. Yuille, (CVPR 2013)

PDF Custom Link

Recent Talks

Handling Occlusion by Cross View Fusion for 3D Human Pose Estimation --- Max Planck Institute for Intelligent Systems

Apr 16, 2020 7:52 AM

Towards Robust 3D Human Pose Estimation in the Wild ---CFCS, Peking University

Apr 10, 2020 7:52 AM