Metraipolis › archive › dc_be873a137a61
GeoLAM: Learning Geometry-Grounded Latent Actions from Unlabeled Human Videos — arXiv:2609.17099
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.17099
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-15T21:30:47.696Z
- SHA-256 of the archived text
33def62eae4f285621f7d53e6cdd115d131d92c8c16ec43bdece239217eabdf7- Stable URL of this page
https://metraipolis.com/documento/dc_be873a137a61
How to cite
GeoLAM: Learning Geometry-Grounded Latent Actions from Unlabeled Human Videos — arXiv:2609.17099 — arXiv (en), https://arxiv.org/abs/2609.17099 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_be873a137a61 (Metraipolis archive dc_be873a137a61, sha256 33def62eae4f285621f7d53e6cdd115d131d92c8c16ec43bdece239217eabdf7)
BibTeX
@misc{dcbe873a137a61,
title = {GeoLAM: Learning Geometry-Grounded Latent Actions from Unlabeled Human Videos — arXiv:2609.17099},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.17099},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_be873a137a61; sha256 33def62eae4f285621f7d53e6cdd115d131d92c8c16ec43bdece239217eabdf7},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.17099. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Human videos provide rich manipulation experience, but extracting action representations that preserve useful motion remains challenging. Visual reconstruction alone can entangle manipulation-related motion with appearance changes and camera movement. We present GeoLAM, a framework for learning geometry-grounded latent actions from action-free human videos. GeoLAM combines future-frame reconstruction through a frozen geometric feature hierarchy with motion supervision from a training-only 4D geometry teacher. The geometric representation provides a structural prior, while the teacher's predictions yield spatially pooled targets capturing 3D displacement, residual image-plane motion, and surface-orientation changes. Visibility and confidence weighting reduces the contribution of unreliable estimates, encouraging continuous latent actions to retain geometric motion without explicit hand-pose or hand-trajectory annotations. After video pretraining without action labels, the learned representation provides transition targets for a world-action model trained on action-labeled robot demonstrations. The model jointly denoises latent actions and executable action chunks, with future-video prediction used only as an auxiliary training task. Deployment therefore requires neither the geometry teacher nor future-video generation. Evaluations on a latent-action benchmark and robotic manipulation tasks demonstrate the strong performance of GeoLAM. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-16 in cs.RO). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Robotics (cs.RO). Pagina del lavoro: https://arxiv.org/abs/2609.17099 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-15T21:30:17.397Z dalla pagina https://arxiv.org/catchup/cs.RO/2026-09-16?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).