Metraipolis › archive › dc_067ebcd66e23

IM-ENGINE: Image Editing for Embodied Data Generation — arXiv:2609.06279

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.06279
License (as archived)
CC0-1.0
Archived
2026-10-16T15:12:39.372Z
SHA-256 of the archived text
6321082d3f96ffb8569f58aec46b9b3f60e84f5fca14b821133708082f63f30a
Stable URL of this page
https://metraipolis.com/documento/dc_067ebcd66e23

How to cite

IM-ENGINE: Image Editing for Embodied Data Generation — arXiv:2609.06279 — arXiv (en), https://arxiv.org/abs/2609.06279 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_067ebcd66e23 (Metraipolis archive dc_067ebcd66e23, sha256 6321082d3f96ffb8569f58aec46b9b3f60e84f5fca14b821133708082f63f30a)
BibTeX
@misc{dc067ebcd66e23,
  title = {IM-ENGINE: Image Editing for Embodied Data Generation — arXiv:2609.06279},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.06279},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_067ebcd66e23; sha256 6321082d3f96ffb8569f58aec46b9b3f60e84f5fca14b821133708082f63f30a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.06279. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Learning-based manipulation requires supervision that is both semantically meaningful and physically executable, but current data pipelines often provide only one of these properties. Human demonstrations capture intent but are costly to collect and constrained by the human-robot embodiment gap, while simulation can scale data generation but often under-specifies functional behavior. We present IM-ENGINE, a simulator-grounded pipeline that uses image editing as an intermediate representation for embodied data generation. Given a rendered scene with known geometry, depth, segmentation, and camera parameters, IM-ENGINE edits the image to inject task-relevant semantics, recovers explicit 3D state using simulator priors and an unchanged anchor object, refines the state in physics, and converts it into robot-executable supervision. We instantiate the pipeline for dexterous grasp synthesis and goal-state generation. For grasping, IM-ENGINE generates a human grasp in image space, recovers the hand-object interaction, retargets it to a robot hand, and refines it into physically validated robot grasps. For goal generation, it edits a rendered scene into a desired outcome, recovers the target-object pose, and refines it into physically valid, semantically meaningful goals and trajectories. This combination of generative semantic priors and simulator grounding enables scalable task-relevant supervision for robot learning.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-09 in cs.CV).
Categorie arXiv: Robotics (cs.RO) ; Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.06279 — 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 2026-10-16T14:49:57.416Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).