|
News
|
|
Research
My research focuses on statistical and learned models of human anatomy, with applications in computer vision, computer graphics, and biomechanics.
I am particularly interested in inferring internal anatomical structures and biomechanically meaningful representations from external observations such as images, body surfaces, and medical data.
|
Publications
|
|
Beyond the Surface: Statistical Approaches to Internal Anatomy Prediction
Marilyn Keller
Ph.D. Thesis, University of Tübingen, November 2024
[Thesis]
[bibtex]
@phdthesis{KellerThesis2024,
title = {Beyond the Surface: Statistical Approaches to Internal Anatomy Prediction},
author = {Keller, Marilyn},
institution = {University of Tübingen},
month = nov,
year = {2024},
month_numeric = {11}
}
My thesis explores how the external body shape can predict internal anatomy, such as skeletal geometry and soft tissue distribution, using statistical methods and datasets derived from medical imaging and biomechanical modeling.
|
|
|
3D Bones: Predicting 3D Bone Locations from External Body Shape
Abdelmouttaleb Dakri,
Vaibhav Arora,
Léo Challier,
Marilyn Keller,
Michael J. Black,
Sergi Pujades
MICCAI 2024
[Paper]
[Supplementary]
[Code]
[Poster]
[Dataset]
[ bibtex]
@inproceedings{Dakri2024:3dbones,
author = {Dakri, Abdelmouttaleb and Arora, Vaibhav and Challier, Léo and Keller, Marilyn and Black, Michael J. and Pujades, Sergi},
title = {{On predicting 3D bone locations inside the human body}},
booktitle = {proceedings of Medical Image Computing and Computer Assisted Intervention -- MICCAI 2024},
year = {2024},
publisher = {Springer Nature Switzerland},
volume = {LNCS 15003},
month = {October},
pages = {336 -- 346}
}
We predict 3D bone locations inside the human body using only the external body surface, leveraging MRI-derived segmented bones for learning correlations between skin and bone location.
|
|
|
HIT: Estimating Internal Human Implicit Tissues from the Body Surface
Marilyn Keller,
Vaibhav Arora,
Abdelmouttaleb Dakri,
Shivam Chandhok,
Jürgen Machann,
Andreas Fritsche,
Michael J. Black,
Sergi Pujades
CVPR 2024
[Paper]
[Supplementary]
[Video]
[Code]
[Poster]
[Dataset]
[ bibtex]
@inproceedings{keller2024hit,
title = {{HIT}: Estimating Internal Human Implicit Tissues from the Body Surface},
author = {Keller, Marilyn and Arora, Vaibhav and Dakri, Abdelmouttaleb and Chandhok, Shivam and Machann, Jürgen and Fritsche, Andreas and Black, Michael J. and Pujades, Sergi},
booktitle = {Proceedings IEEE/CVF Conf.~on Computer Vision and Pattern Recognition (CVPR)},
month = jun,
year = {2024},
month_numeric = {6},
pages = {3480--3490}
}
We leverage the shape of the body surface to infer the 3D location of subcutaneous adipose tissue, lean tissue, and long bones.
|
|
From Skin to Skeleton: Towards Biomechanically Accurate 3D Digital Humans
Marilyn Keller,
Keenon Werling,
Soyong Shin,
Scott Delp,
Sergi Pujades,
C. Karen Liu,
Michael J. Black
SIGGRAPH ASIA 2023
Honorable Mention for Best Paper
[Paper]
[Supplementary]
[Code]
[ bibtex]
@article{keller2023skel,
title={From skin to skeleton: Towards biomechanically accurate 3d digital humans},
author={Keller, Marilyn and Werling, Keenon and Shin, Soyong and Delp, Scott and Pujades, Sergi and Liu, C Karen and Black, Michael J},
journal={ACM Transactions on Graphics (TOG)},
volume={42},
number={6},
pages={1--12},
year={2023},
publisher={ACM New York, NY, USA}
}
We learn SKEL, a model that upgrades SMPL with more anatomical degrees of freeedom.
|
|
OSSO: Obtaining Skeletal Shape from Outside
Marilyn Keller,
Silvia Zuffi,
Michael J. Black,
Sergi Pujades
CVPR 2022
[Paper]
[Supplementary]
[Video]
[Code]
[Poster]
[ bibtex]
@InProceedings{keller2022osso,
title = {{OSSO}: Obtaining Skeletal Shape from Outside},
author = {Keller, Marilyn and Zuffi, Silvia and Black, Michael J. and Pujades, Sergi},
booktitle = {Proceedings IEEE/CVF Conf.~on Computer Vision and Pattern Recognition (CVPR)},
month = jun,
year = {2022},
month_numeric = {6},
pages = {20492-20501}
}
We learn to infer the skeleton mesh from the external appearance.
|
|
Optimizing the 3D Plate Shape for Proximal Humerus Fractures
Marilyn Keller,
Marcell Krall, James Smith, Hans Clement, Alexander M. Kerner,
Andreas Gradischar, Ute Schäfer, Michael J. Black,
Sergi Pujades
MICCAI 2023
[Paper]
[Supplementary]
[Poster]
[Code]
[ bibtex]
@inproceedings{keller2023humerusplate,
title = {Optimizing the {3D} Plate Shape for Proximal Humerus Fractures},
author = {Keller, Marilyn and Krall, Marcell and Smith, James and Clement, Hans and Kerner, Alexander M.
and Gradischar, Andreas and Sch{ä}fer, Ute and Black, Michael J and Weinberg, Annelie and Pujades, Sergi},
booktitle = {International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI)},
pages = {487--496},
month = oct,
year = {2023},
}
We present a method to extract custom implant 3D shapes and generate sets of implant shapes.
|
|
|
Learning a statistical full spine model from partial observations
Di Meng,
Marilyn Keller,
Edmond Boyer,
Michael J. Black,
Sergi Pujades
Shape in Medical Imaging
[Paper]
[Code]
[ bibtex]
@inproceedings{meng2020spinemodel,
title = {Learning a statistical full spine model from partial observations},
author = {Meng, Di and Keller, Marilyn and Boyer, Edmond and Black, Michael and Pujades, Sergi},
booktitle = {Shape in Medical Imaging},
pages = {122--133},
month = oct,
year = {2020},
doi = {10.1007/978-3-030-61056-2_10},
month_numeric = {10}
}
Learning a statisticcal Spine model from partial observations. |
|
|
Obstacles Awareness Methods from Occupancy Map for Free Walking in VR
Marilyn Keller,
Tristan Tchilinguirian
IEEE Conference on Virtual Reality and 3D User Interfaces (VR) 2019
[Paper]
[Poster]
[ bibtex]
@InProceedings{KELLER:IEEEVR:2019,
title = {Obstacles Awareness Methods from Occupancy Map for Free Walking in VR},
author = {Keller, Marilyn and Tchilinguirian, Tristan},
booktitle = {Proceedings IEEE Conference on Virtual Reality and 3D User Interfaces (VR)},
pages = {1012-1013},
publisher = {IEEE},
month = mar,
year = {2019},
doi = {10.1109/VR.2019.8798260},
month_numeric = {3}
}
Evaluating and comparing different visualization methods to enable real walking in Virtual Reality.
|
|
|
Game room map integration in virtual environments for free walking
Marilyn Keller,
Frédéric Exposito
Proceedings VR 2018
[Paper]
[ bibtex]
@proceedings{KELLER:IEEEVR:2018,
title = {Game room map integration in virtual environments for free walking},
author = {Keller, Marilyn and Exposito, Frédéric},
pages = {763-764},
year = {2018},
doi = {10.1109/VR.2018.8446385}
}
We demonstrate a method to enable free walking with an HMD by showing the position of the real obstacles to the user.
|
|
|
Setup of a capture system consisting of multiple
video and depth cameras
Marilyn Keller
Master Thesis
[Thesis]
[ bibtex]
@masterthesis{keller:masterthesis:multicam,
title = {Setup of a capture system consisting of multiple
video and depth cameras},
author = {Marilyn Keller},
year = 2016,
month = {August},
school = {Karlsruhe Institute of Technology (KIT)}},
Setting up a real time multi camera setup, made from several color and depth camera. We encode the scene in a compact way and render it in unreal engine.
|
Miscellaneous
Some other unpublished work:
|
GauMarmoset: 4DGS reconstruction of common marmoset (Poster)
Kaho Oshita, Marilyn Keller, Takaaki Kaneko, Takako Miyabe, Ken Nakae, Diego Thomas
MIRU 2026
[Poster]
We propose an automated 4D Gaussian Splatting framework that reconstructs individualized dynamic 3D shape and appearance for common marmosets from multi-view external videos, providing a scalable foundation for digital twins.
|
|
4D capture and analysis of Kyudo shooting motion using 2-views RGB-D cameras (Poster)
Kaho Oshita, Marilyn Keller, Diego Thomas
MIRU 2025
[Poster]
We present a 4D capture and analysis system for studying Kyudo shooting motion using two-view RGB-D cameras.
|
|
