- [Jan 25, 2026] Workshop website launched, with Call-for-Papers and speakers announced.
- Interested in our previous editions!
- CVPR 2025 @ Nashville: HuMoGen2025
- CVPR 2024 @ Seattle: HuMoGen2024
Introduction
Motion is one of the fundamental attributes of human life, shaping our actions, behavior, and social interactions. Over the past years, generative motion modeling has gained a surge of interest for the CVPR community and has facilitated a variety of applications such as embodied intelligence, robotics learning, 3D character animations, AR/VR, crowd simulation, sports analytics, and many more.
As the field evolves rapidly with emerging topics and techniques, we propose the third edition of the workshop on Human Motion Generation (HuMoGen) with new perspectives on simulation, animation, and VR applications. Through this workshop, we intend to continue bringing the human motion synthesis community together to foster discussions about existing challenges and future directions, specifically focusing on the gap between academic and industrial perspectives.
Invited Speakers
 |
 |
Sebastian Starke |
Libin Liu |
| Meta Reality Labs |
Peking University |
 |
 |
Alexander Richard |
Christian Theobalt |
| Reality Labs Research @ Meta |
Max-Planck Institute for Informatics Saarland University Saarbruecken Center for Visual Computing |
Call for Papers
We invite submissions of full-length papers (8 pages excluding the references) for workshop proceedings. The topics covered in the workshop include but are not limited to:
- Advances in generative human motion synthesis models
- Advances in motion synthesis in VR/AR/XR applications
- Advances in motion synthesis for robotic learning
- Advances in character animation
- Advances in controllable motion synthesis, e.g., text, audio, prefix, trajectories, key poses
- Advances in co-speech gesture synthesis
- Advances in interaction motion synthesis include human-scene, humanhuman, and human-object interactions
- Advances in physically-plausible motion synthesis
- Evaluation metrics for motion synthesis literature
- Ethical considerations in motion synthesis and dataset bias
Submission Instructions
All submissions should follow the CVPR 2026 instructions. The papers will be subject to a double-blind review process, i.e. authors must not identify themselves on the submitted papers. The reviewing process is single-stage without rebuttals.
- Online Submission System: [Under Construction].
- Submission Format: official CVPR template (double column; no more than 8 pages, excluding reference).
All authors submitting a paper are required to have an OpenReview profile. New profiles with institutional emails are automatically activated, while those without one undergo a moderation process, taking up to two weeks.
Call for 1-page abstracts
Conference and Journal accepted papers from the areas above are invited to submit one-page abstract via humogencvpr@gmail.com, including title, publish veune, author information, summary of research, teaser figure and webpage (optional but recommended). Accepted abstracts will be presented at the workshop's posters session. The deadline is May 10, yet notifications will be asynchronous - approximately one week from submission.
Timeline Table (11:59 PM, Pacific Time)
- Full-paper submission deadline: March 7, 2026
- Notification to authors: March 19, 2026
- Camera-ready deadline: April 10, 2026
- 1-page Abstract submission deadline: May 10, 2026
Workshop Organizers
 |
 |  |
Peizhuo Li |
Rishabh Dabral |
Sigal Raab |
| ETH Zurich |
Max-Planck Institute for Informatics |
Weizmann Institute of Science |
 |
 |
 |
Guy Tevet |
Chuan Guo |
Yuxuan Mu |
| Stanford University |
Meta Reality Labs |
Simon Fraser University |
 |
 |
|
Ikhsanul Habibie |
Amit Haim Bermano |
Christian Theobalt |
| Max-Planck Institute for Informatics |
Tel-Aviv University |
Max-Planck Institute for Informatics Saarland University |
Contact Info
E-mail: humogencvpr@gmail.com
Acknowledgement
Website template borrowed from: https://rhobin-challenge.github.io/