Vasileios Choutas

I am a Ph.D. student at the Perceiving Systems department of the Max Planck Institute for Intelligent Systems, supervised by Dimitris Tzionas and Michael J. Black. I am also part of the Max Planck ETH Center for Learning Systems.

During the summer of 2022 I was an intern at Meta Reality Labs Research, working with Kaiwen Guo. From June to November 2021 I was an intern at the Mixed Reality & AI Lab in Zurich, working with Federica Bogo and Julien Valentin.

Prior to joining PS, I was fortunate to visit and work with Matthias Nießner, at the Visual Computing & Artificial Intelligence. Before going to Germany, I spent 8 wonderful months in Grenoble, working with Philippe Weinzaepfel, Jérome Revaud and Cordelia Schmid, as a member of the Thoth team and NaverLabs Europe. This journey started at the Aristotle University of Thessaloniki , where I studied Electrical and Computer Engineering.

profile photo
News
  • October 2022: TAAD won the MultiSports spatio-temporal action detection challenge at ECCV'22.
  • July 4, 2022: I started my internship at Meta Reality Labs Research.
  • SHAPY was a best paper finalist at CVPR 2022.
  • I was recognized as an Outstanding Reviewer at 3DV 2021.
  • I was recognized as an Outstanding Reviewer at CVPR 2021.
  • June 1, 2021: I started my internship at the Microsoft Mixed Reality & AI Lab in Zurich.
Publications
Spatio-Temporal Action Detection Under Large Motion
Gurkirt Singh, Vasileios Choutas, Suman Saha, Fisher Yu, Luc Van Gool
Winter Conference on Applications of Computer Vision (WACV), 2023
[PDF] [arxiv] [slides] [challenge]
[bibtex]
@InProceedings{Singh_2023_WACV, author = {Singh, Gurkirt and Choutas, Vasileios and Saha, Suman and Yu, Fisher and Van Gool, Luc}, title = {Spatio-Temporal Action Detection Under Large Motion}, booktitle = {Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)}, month = {January}, year = {2023}, pages = {6009-6018} }

Use tracking to improve action detection.

Learning to Fit Morphable Models
Vasileios Choutas, Federica Bogo, Jingjing Shen Julien Valentin
European Conference on Computer Vision (ECCV), 2022
[project page] [PDF] [arxiv] [poster] [video]
[bibtex]
@inproceedings{Choutas:ECCV:2022,  author = {Vasileios Choutas and Federica Bogo and Jingjing Shen and Julien Valentin},  title = {Learning to Fit Morphable Models},  booktitle = {European Conference on Computer Vision (ECCV)},  month = {October},  year = {2022}, }

Neural optimizer inspired from Levenberg-Marquadt.

Accurate 3D Body Shape Regression using Metric and Semantic Attributes
Vasileios Choutas*, Lea Müller*, Chun-Hao Paul Huang, Siyu Tang, Dimitris Tzionas, Michael J. Black
Computer Vision and Pattern Recognition (CVPR), 2022
Oral Presentation
Best paper candidate
[project page] [PDF] [arxiv] [code] [video]
[bibtex]
@inproceedings{Shapy:CVPR:2022,  author = {Choutas, Vasileios and M\"uller, Lea and Huang, Chun-Hao P. and Tang, Siyu and Tzionas, Dimitrios and Black, Michael J.},  title = {Accurate 3D Body Shape Regression Using Metric and Semantic Attributes},  booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},  month = {June},  year = {2022},  pages = {2718-2728} }

Linguistic attribute scores and anthropometric body measurements are effective proxies for 3D body shape supervision.

GOAL: Generating 4D Whole-Body Motion for Hand-Object Grasping
Omid Taheri, Vasileios Choutas, Michael J. Black, Dimitris Tzionas
Computer Vision and Pattern Recognition (CVPR), 2022
[project page] [PDF ] [arxiv] [code]
[bibtex]
@inproceedings{Taheri_2022_CVPR,  author = {Taheri, Omid and Choutas, Vasileios and Black, Michael J. and Tzionas, Dimitrios},  title = {GOAL: Generating 4D Whole-Body Motion for Hand-Object Grasping},  booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},  month = {June},  year = {2022},  pages = {13263-13273} }

Generate grasping motion for a target object.

Collaborative Regression of Expressive Bodies using Moderation
Yao Feng *, Vasileios Choutas*, Timo Bolkart, Dimitris Tzionas, Michael J. Black
International Conference on 3D Vision, 2021
[project page] [PDF] [supplementary] [code]
[bibtex]
@inproceedings{PIXIE:2021,  title={Collaborative Regression of Expressive Bodies using Moderation},  author={Yao Feng and Vasileios Choutas and Timo Bolkart and Dimitrios Tzionas and Michael J. Black},  booktitle={International Conference on 3D Vision (3DV)},  year={2021},  pages={792-804}, }

Reconstruct expressive 3D humans from a single RGB image by adaptively aggregating information from part experts.

Monocular Expressive Body Regression through Body-Driven Attention
Vasileios Choutas, Georgios Pavlakos, Timo Bolkart, Dimitris Tzionas, Michael J. Black
European Conference on Computer Vision (ECCV), 2020
[project page] [PDF] [supplementary] [arxiv] [Long video] [Short video] [code]
[bibtex]
@inproceedings{ExPose:2020,  title = {Monocular Expressive Body Regression through Body-Driven Attention},  author = {Choutas, Vasileios and Pavlakos, Georgios and Bolkart, Timo and Tzionas, Dimitrios and Black, Michael J.},  booktitle = {European Conference on Computer Vision (ECCV)},  volume = {LNCS 12355},  pages = {20--40},  year = {2020} }

Regression-based expressive capture of 3D humans from a single RGB image.

Resolving 3D Human Pose Ambiguities with 3D Scene Constraints
Mohamed Hassan, Vasileios Choutas, Dimitris Tzionas, Michael J. Black
International Conference on Computer Vision (ICCV), 2019
[project page] [PDF] [arxiv] [video] [code]
[bibtex]
@inproceedings{PROX:2019,  title = {Resolving {3D} Human Pose Ambiguities with {3D} Scene Constraints},  author = {Hassan, Mohamed and Choutas, Vasileios and Tzionas, Dimitrios and Black, Michael J.},  booktitle = {Proceedings International Conference on Computer Vision},  pages = {2282--2292},  publisher = {IEEE},  month = oct,  year = {2019},  url = {https://prox.is.tue.mpg.de},  month_numeric = {10} }

Leveraging scene constraints to improve 3D human pose and shape estimation

Expressive Body Capture: 3D Hands, Face and Body from a Single Image
Georgios Pavlakos*, Vasileios Choutas*, Nima Ghorbani, Ahmed A. A. Osman, Timo Bolkart, Dimitris Tzionas, Michael J. Black
Computer Vision and Pattern Recognition (CVPR), 2019
Oral Presentation
[project page] [PDF] [supplementary] [arxiv] [video] [poster] [code]
[bibtex]
@inproceedings{SMPL-X:2019,  title = {Expressive Body Capture: 3D Hands, Face, and Body from a Single Image},  author = {Pavlakos, Georgios and Choutas, Vasileios and Ghorbani, Nima and Bolkart, Timo and Osman, Ahmed A. A. and Tzionas, Dimitrios and Black, Michael J.},  booktitle = {Proceedings IEEE Conf. on Computer Vision and Pattern Recognition (CVPR)},  pages = {10975--10985},  month = jun,  year = {2019},  url = {http://smpl-x.is.tue.mpg.de},  month_numeric = {6} }

Expressive capture of bodies, hands and faces from a single RGB image.

PoTion: Pose MoTion Representation for Action Recognition
Vasileios Choutas, Philippe Weinzaepfel, Jérome Revaud and Cordelia Schmid
Computer Vision and Pattern Recognition (CVPR), 2018
[project page] [PDF]
[bibtex]
@inproceedings{Choutas_2018_CVPR,  author = {Choutas, Vasileios and Weinzaepfel, Philippe and Revaud, Jérôme and Schmid, Cordelia},  title = {PoTion: Pose MoTion Representation for Action Recognition},  booktitle = {Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},  month = {June},  year = {2018}  pages = {7024-7033}, }

A representation that encodes human motion in videos.


Special thanks to Jon Barron for the template