WELCOME TO THE DIGITAL HUMANS & 4D CAPTURE LAB ! Lab Logo
( part of the Computer Vision & Graphics Lab )

Our societies are constently evolving, adapting to new technologies and needs. In the new era of technologies powered by AI, how to represent humans in the digital worlds is a key question. To answer this question we focus our research on creating digital humans with real shape, appearance, motions and ability to interact naturally with the environment. To leverage all the power of AI for 3D vision we also work in efficient 4D capture systems for humans but also for the cities, with exciting applications in various domains such as architectural design.


Latest news

Paper accepted at WACV 2025! "VortSDF"

Paper accepted at WACV 2025! "Weakly-Supervised 3D Reconstruction of Clothed Humans via Normal Maps"

Paper accepted at SIGGRAPH ASIA 2024!

Paper accepted at ACCV 2024!

Paper accepted at IROS 2024!

Best paper award at MIRU 2024 (MIRU Nagao Award (MIRU長尾賞))!

MIRU Excellent Paper Award (MIRU優秀賞) at MIRU 2024!

Dr.Benjamin Renoust (Median Technologies) gave a talk onClinical data and AI on November 1st 2023!

Prof. Vincent Nozick gave a talk on Geometric Algebra!

Prof. Vincent Nozick, from Gustave Eiffel University (France) will visit the lab from April 10th to April 14th 2023!

Prof. Joao Paulo do Monte Lima, from Univ. Pernambuco will visit the lab from March 4th to May 4th 2023!


Go to News Section
Latest publications

Diego Thomas (Kyushu University), Briac Toussaint (INRIA Grenoble), Jean Sebastien Franco (INRIA Grenoble) and Edmond Boyer (Meta Reality Lab). "VortSDF: 3D Modeling with Centroidal Voronoi Tesselation on Signed Distance Field" In arXiv, 2023.


Jane Wu (Stanford), Diego Thomas (Kyushu University) and Ronald Fedkiw (Stanford, Epic Games). "Weakly-Supervised 3D Reconstruction of Clothed Humans via Normal Maps" In arXiv, 2023.


Teshima, Hitoshi, Naoki Wake, Diego Thomas, Yuta Nakashima, Hiroshi Kawasaki, and Katsushi Ikeuchi. "Deep Gesture Generation for Social Robots Using Type-Specific Libraries." In 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 8286-8291. IEEE, 2022.



Go to Publi Section
Latest projects

Better neural implicit surface reconstruction with Centroidal Voronoi Tessellation [2022 ~]


Animatable avatars [2021 ~]


Weakly supervised 3D clothed human bodies from few images [2021 ~]


Supporting architectural simulations and visualizations with 3D vision [2021 ~]


Aerial drone-based RGB-D mapping [2020 ~]

Kyushu University Logo