My previous works center on the generative modeling of human motion, behavior, and interaction — from monocular motion capture and real-time character control to today’s large-scale, multimodal generative models. The projects and publications below trace this line of work over the years.

I also serve as a reviewer for conferences and journals including: SIGGRAPH, SIGGRAPH Asia, Eurographics, ICCV, CVPR, ICLR, AAAI; TOG, TVCG, CGF, T-PAMI, TIP, and others.

Projects

LPM 1.0 (Large Performance Model)

LPM 1.0 generates real-time video with full-duplex conversation, identity-consistent infinite-length generation, and nuanced human-like performance.

LPM Team, 2026

[webpage] [paper]

Pre-print

MotionPersona: Characteristics-aware Locomotion Control

Mingyi Shi, Wei Liu, Jidong Mei, Wangpok Tse, Rui Chen, Xuelin Chen, Taku Komura

[webpage] [arXiv] [Unity Demo]

arXiv 2025

It Takes Two: Real-time Co-Speech Two-person's Interaction Generation via Reactive Auto-regressive Diffusion Model

Mingyi Shi*, Dafei Qin*, Leo Ho, Zhouyingcheng Liao, Yinghao Huang, Junichi Yamagishi, Taku Komura
*Joint first author

[arXiv]

arXiv 2024

Prior-First, Condition-Second: Scalable and Controllable Hand Motion Completion

Mingyi Shi, Xuelin Chen, Taku Komura

[webpage] [paper] [code] [blender]

ACM SIGGRAPH / Eurographics Symposium on Computer Animation 2026, Computer Graphics Forum, Best Paper Honorable Mention Award

🎧 MOSPA: Spatial Audio-Driven Human Motion Generation

Shuyang Xu, Zhiyang Dou*#, Mingyi Shi, Liang Pan, Leo Ho, Jingbo Wang, Yuan Liu, Cheng Lin, Yuexin Ma, Wenping Wang#, Taku Komura#

[webpage] [paper] [video] [code]

NeurIPS 2025, Spotlight

InterAct: A Large-Scale Dataset of Dynamic, Expressive and Interactive Activities between Two People in Daily Scenarios

Leo Ho*, Yinghao Huang*, Dafei Qin, Mingyi Shi, Wangpok Tse, Wei Liu, Junichi Yamagishi, Taku Komura
* Joint first author

[webpage] [paper] [video] [dataset]

ACM SIGGRAPH / Eurographics Symposium on Computer Animation 2025, ACM on Computer Graphics and Interactive Techniques

Taming Diffusion Probabilistic Models for Character Control

Rui Chen*, Mingyi Shi*, Shaoli Huang, Ping Tan, Taku Komura, Xuelin Chen
* Joint first author

[webpage] [paper] [code]

SIGGRAPH 2024, Featured in Technical Papers Video Trailer!

PhaseMP: Robust 3D Pose Estimation via Phase-conditioned Human Motion Prior

Mingyi Shi, Sebastian Starke, Yuting Ye, Taku Komura, Jungdam Won

[webpage] [paper] [code]

ICCV 2023

Normalized Human Pose Features for Human Action Video Alignment

Jingyuan Liu, Mingyi Shi, Qifeng Chen, Hongbo Fu, Chiew-Lan Tai

[paper] [supp]

ICCV 2021, Oral

MotioNet: 3D Human Motion Reconstruction from Monocular Video with Skeleton Consistency

Mingyi Shi, Kfir Aberman, Andreas Aristidou, Taku Komura, Dani Lischinski, Daniel Cohen-Or, Baoquan Chen

[webpage] [paper] [video] [code]

SIGGRAPH Asia 2020, ACM Transactions on Graphics

Deep Video-Based Performance Cloning

Kfir Aberman, Mingyi Shi, Jing Liao, Dani Lischinski, Baoquan Chen, Daniel Cohen-Or

[paper] [video]

Eurographics 2019, Computer Graphics Forum

Neural Best-Buddies: Sparse Cross-Domain Correspondence

Kfir Aberman, Jing Liao, Mingyi Shi, Dani Lischinski, Baoquan Chen, Daniel Cohen-Or

[webpage] [paper] [video] [code]

SIGGRAPH 2018, ACM Transactions on Graphics