Xinyu Yi

Xinyu Yi

I'm a Research Scientist at ByteDance, focusing on robotics research. Previously, I obtained my Ph.D. from Tsinghua University in 2025, supervised by Prof. Feng Xu. Before that, I received my B.S. from School of the Gifted Young, University of Science and Technology of China in 2020.

My research aims to build physics-based human motion capture and human-environment interaction systems. During my Ph.D., I worked extensively on real-time human motion capture from sparse lightweight body-worn sensors.

Research

DiffCap
DiffCap: Diffusion-based Real-time Human Motion Capture using Sparse IMUs and a Monocular Camera

Shaohua Pan, Xinyu Yi, Yan Zhou, Weihua Jian, Yuan Zhang, Pengfei Wan, Feng Xu
Transactions on Visualization and Computer Graphics (TVCG), 2025

MagShield
MagShield: Towards Better Robustness in Sparse Inertial Motion Capture Under Magnetic Disturbances

Yunzhe Shao, Xinyu Yi, Lu Yin, Shihui Guo, Junhai Yong, Feng Xu
International Conference on Computer Vision (ICCV), 2025

BaroPoser
BaroPoser: Real-time Human Motion Tracking from IMUs and Barometers in Everyday Devices

Libo Zhang, Xinyu Yi, Feng Xu
ACM Symposium on User Interface Software and Technology (UIST), 2025

GlobalPose
Improving Global Motion Estimation in Sparse IMU-based Motion Capture with Physics

Xinyu Yi, Shaohua Pan, Feng Xu
ACM Transactions on Graphics (Proc. of SIGGRAPH), 2025

TIC
Transformer IMU Calibrator: Dynamic On-body IMU Calibration for Inertial Motion Capture

Chengxu Zuo, Jiawei Huang, Xiao Jiang, Yuan Yao, Xiangren Shi, Rui Cao, Xinyu Yi, Feng Xu, Shihui Guo, Yipeng Qin
ACM Transactions on Graphics (Proc. of SIGGRAPH), 2025 (Best Paper Award)

PNP
Physical Non-inertial Poser (PNP): Modeling Non-inertial Effects in Sparse-inertial Human Motion Capture

Xinyu Yi, Yuxiao Zhou, Feng Xu
Proc. of SIGGRAPH, 2024

HOIC
Hand-Object Interaction Controller (HOIC): Deep Reinforcement Learning for Reconstructing Interactions with Physics

Haoyu Hu, Xinyu Yi, Zhe Cao, Jun-Hai Yong, Feng Xu
Proc. of SIGGRAPH, 2024

LIP
Loose Inertial Poser: Motion Capture with IMU-attached Loose-Wear Jacket

Chengxu Zuo, Yiming Wang, Lishuang Zhan, Shihui Guo, Xinyu Yi, Feng Xu, Yipeng Qin
Computer Vision and Pattern Recognition (CVPR), 2024

RobustCap
Fusing Monocular Images and Sparse IMU Signals for Real-time Human Motion Capture

Shaohua Pan, Qi Ma, Xinyu Yi, Weifeng Hu, Xiong Wang, Xingkang Zhou, Jijunnan Li, Feng Xu
Proc. of SIGGRAPH Asia, 2023

EgoLocate
EgoLocate: Real-time Motion Capture, Localization, and Mapping with Sparse Body-mounted Sensors

Xinyu Yi, Yuxiao Zhou, Marc Habermann, Vladislav Golyanik, Shaohua Pan, Christian Theobalt, Feng Xu
ACM Transactions on Graphics (Proc. of SIGGRAPH), 2023

PIP
Physical Inertial Poser (PIP): Physics-aware Real-time Human Motion Tracking from Sparse Inertial Sensors

Xinyu Yi, Yuxiao Zhou, Marc Habermann, Soshi Shimada, Vladislav Golyanik, Christian Theobalt, Feng Xu
Computer Vision and Pattern Recognition (CVPR), 2022 (Best Paper Finalist)

PhysInteraction
Physical Interaction: Reconstructing Hand-object Interactions with Physics

Haoyu Hu, Xinyu Yi, Hao Zhang, Jun-Hai Yong, Feng Xu
Proc. of SIGGRAPH Asia, 2022

TransPose
TransPose: Real-time 3D Human Translation and Pose Estimation with Six Inertial Sensors

Xinyu Yi, Yuxiao Zhou, Feng Xu
ACM Transactions on Graphics (Proc. of SIGGRAPH), 2021

HierarIK
HierarIK: Hierarchical Inverse Kinematics Solver for Human Body and Hand Pose Estimation

Xinyu Yi, Yuxiao Zhou, Feng Xu
CAAI International Conference on Artificial Intelligence (CICAI), 2021 (Best Student Paper Finalist)

Education

Ph.D.   Sep 2020 - July 2025
School of Software, Tsinghua University
Advisor: Prof. Feng Xu
B.S.   Sep 2016 - July 2020
University of Science and Technology of China
School of the Gifted Young

Honors & Awards