MoFM: A Large-Scale Human Motion Foundation Model
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915172037165056 |
|---|---|
| author | Baharani, Mohammadreza Noghre, Ghazal Alinezhad Pazho, Armin Danesh Maldonado, Gabriel Tabkhi, Hamed |
| author_facet | Baharani, Mohammadreza Noghre, Ghazal Alinezhad Pazho, Armin Danesh Maldonado, Gabriel Tabkhi, Hamed |
| contents | Foundation Models (FM) have increasingly drawn the attention of researchers due to their scalability and generalization across diverse tasks. Inspired by the success of FMs and the principles that have driven advancements in Large Language Models (LLMs), we introduce MoFM as a novel Motion Foundation Model. MoFM is designed for the semantic understanding of complex human motions in both time and space. To facilitate large-scale training, MotionBook, a comprehensive human motion dictionary of discretized motions is designed and employed. MotionBook utilizes Thermal Cubes to capture spatio-temporal motion heatmaps, applying principles from discrete variational models to encode human movements into discrete units for a more efficient and scalable representation. MoFM, trained on a large corpus of motion data, provides a foundational backbone adaptable to diverse downstream tasks, supporting paradigms such as one-shot, unsupervised, and supervised tasks. This versatility makes MoFM well-suited for a wide range of motion-based applications. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_05432 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MoFM: A Large-Scale Human Motion Foundation Model Baharani, Mohammadreza Noghre, Ghazal Alinezhad Pazho, Armin Danesh Maldonado, Gabriel Tabkhi, Hamed Computer Vision and Pattern Recognition Machine Learning Foundation Models (FM) have increasingly drawn the attention of researchers due to their scalability and generalization across diverse tasks. Inspired by the success of FMs and the principles that have driven advancements in Large Language Models (LLMs), we introduce MoFM as a novel Motion Foundation Model. MoFM is designed for the semantic understanding of complex human motions in both time and space. To facilitate large-scale training, MotionBook, a comprehensive human motion dictionary of discretized motions is designed and employed. MotionBook utilizes Thermal Cubes to capture spatio-temporal motion heatmaps, applying principles from discrete variational models to encode human movements into discrete units for a more efficient and scalable representation. MoFM, trained on a large corpus of motion data, provides a foundational backbone adaptable to diverse downstream tasks, supporting paradigms such as one-shot, unsupervised, and supervised tasks. This versatility makes MoFM well-suited for a wide range of motion-based applications. |
| title | MoFM: A Large-Scale Human Motion Foundation Model |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2502.05432 |