Tairan Yin

PhD in Computer Science

About me

I am committed to advancing virtual reality technology, and my current research interests are in building realistic and interactive virtual humans in real-time in virtual reality environments to enhance the user's immersive experience. My research interests also cover computer graphics, computer animation, computer vision and crowd simulation.

Experience

Inria

France

Early Stage Researcher

Nov. 2020 - Apr. 2024

Funded by the European Union's Marie Skłodowska-Curie Actions grant under the CLIPE-ITN project, my PhD research focused on prooposing a novel crowd motion data collection paradigm using Virtual Reality. This research experiences consisted of designing and implementing VR environments, performing user experiment, the use of VR and motion capture hardware, the use of crowd simulation algorithm such as RVO and ORCA, data analysis, and scientific writing and publication.

Thesis:

The One-Man-Crowd: Towards Single-User Capture of Collective Motions using Virtual Reality [PDF] [Defense]

Experience gained:

Programing: Unity, C#, Matlab, C++, OpenCV, Python.

Technique: Xsens, Pimax, HTC Vive, HP VR backpacks

Others: scientific writing, writing of ethic proposal.


Visiting Researcher

Nov. 2023

This visit happened within the CLIPE-ITN project at the end of my PhD. Hosted by my co-supervisor, the visit was mainly about making echanges about VR-experiments and user-studies.

LIX

France


Golaem

France

Visiting Research Intern

Nov. 2022 - Jan. 2023

This visit happened within the CLIPE-ITN project. A first-step exploration of how the One-Man-Crowd dataset can serve motion generation and motion style transfer. We trained a simple auto-encoder with 6 people’s OMC data and successfully tested it on 2 unseen people’s data.

Experience gained:

Programing: Python, PyTorch


Visiting Research Intern

Feb. 2022 - May. 2022

This visit happened within the CLIPE-ITN project. The goal of this visit is to setup a VR Motion Capture solution for the department’s amazing capture hull that record and replay human motion directly through the SMPLX model.

Experience gained:

Technique: Vicon Motion Capture System.

MPI-IS

Germany


SKLVRTS

China

Research Assistant

Sep. 2019 - Jun. 2022

Under my master's research project, this experience happened at the State Key Laboratory of Virtual Reality technology and System. The reserach topic is reconstructing human hand shape from depth map input using a graph convolutional network (GCN). The network contains a stacked hourglass network to predict hand joints on the depth map, and a GCN to predict the hand shape.

Experience gained:

Programing: C++, OpenGL, OpenCV, CUDA, Python, PyTorch

Technique: Depth cameras such as Azure Kinect and Intel Realsense

Publications

T. Yin, L. Hoyet, M. Christie, M-P. Cani, J. Pettré

With or Without You: Effect of Contextual and Responsive Crowds on VR-based Crowd Motion Capture [Paper] [Video]

IEEE TVCG (Honorable mention for the best journal papers award at IEEE VR 2024)

T. Yin, L. Hoyet, M. Christie, M-P. Cani, J. Pettré

The One-Man-Crowd: Single User Generation of Crowd Motions Using Virtual Reality [Paper] [Video]

IEEE TVCG (Nominee for the best journal paper award at IEEE VR 2022)

Education

PhD in Computer Science - 2024

University of Rennes, France


M.Eng. in System Engineer - 2020

Beihang University


Ingénieur Généraliste - 2020

Ecole Centrale de Marseille (Now became Ecole Centrale Méditerranée)


B.Sc in Mathematics and Applied Mathematics - 2017

Beihang University

Buildings 1

Buildings 2

Nature

Playground

A simple PDF editor

A PDF web editor open sourced by ShizukuIchi on Github. I deployed a copy on my own server for test purpose. This will be later removed.

Get Current Location

This is a small and preliminary test function for an upcoming project that aims to support touristic aurora observation. This will be removed when a further step is ready.

Your location will appear here.