EgoBridge: Domain Adaptation for Generalizable Imitation from Egocentric Human Data

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Punamiya, Ryan, Patel, Dhruv, Aphiwetsa, Patcharapong, Kuppili, Pranav, Zhu, Lawrence Y., Kareer, Simar, Hoffman, Judy, Xu, Danfei
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866908556497780736
author Punamiya, Ryan
Patel, Dhruv
Aphiwetsa, Patcharapong
Kuppili, Pranav
Zhu, Lawrence Y.
Kareer, Simar
Hoffman, Judy
Xu, Danfei
author_facet Punamiya, Ryan
Patel, Dhruv
Aphiwetsa, Patcharapong
Kuppili, Pranav
Zhu, Lawrence Y.
Kareer, Simar
Hoffman, Judy
Xu, Danfei
contents Egocentric human experience data presents a vast resource for scaling up end-to-end imitation learning for robotic manipulation. However, significant domain gaps in visual appearance, sensor modalities, and kinematics between human and robot impede knowledge transfer. This paper presents EgoBridge, a unified co-training framework that explicitly aligns the policy latent spaces between human and robot data using domain adaptation. Through a measure of discrepancy on the joint policy latent features and actions based on Optimal Transport (OT), we learn observation representations that not only align between the human and robot domain but also preserve the action-relevant information critical for policy learning. EgoBridge achieves a significant absolute policy success rate improvement by 44% over human-augmented cross-embodiment baselines in three real-world single-arm and bimanual manipulation tasks. EgoBridge also generalizes to new objects, scenes, and tasks seen only in human data, where baselines fail entirely. Videos and additional information can be found at https://ego-bridge.github.io
format Preprint
id arxiv_https___arxiv_org_abs_2509_19626
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle EgoBridge: Domain Adaptation for Generalizable Imitation from Egocentric Human Data
Punamiya, Ryan
Patel, Dhruv
Aphiwetsa, Patcharapong
Kuppili, Pranav
Zhu, Lawrence Y.
Kareer, Simar
Hoffman, Judy
Xu, Danfei
Robotics
Computer Vision and Pattern Recognition
Machine Learning
Egocentric human experience data presents a vast resource for scaling up end-to-end imitation learning for robotic manipulation. However, significant domain gaps in visual appearance, sensor modalities, and kinematics between human and robot impede knowledge transfer. This paper presents EgoBridge, a unified co-training framework that explicitly aligns the policy latent spaces between human and robot data using domain adaptation. Through a measure of discrepancy on the joint policy latent features and actions based on Optimal Transport (OT), we learn observation representations that not only align between the human and robot domain but also preserve the action-relevant information critical for policy learning. EgoBridge achieves a significant absolute policy success rate improvement by 44% over human-augmented cross-embodiment baselines in three real-world single-arm and bimanual manipulation tasks. EgoBridge also generalizes to new objects, scenes, and tasks seen only in human data, where baselines fail entirely. Videos and additional information can be found at https://ego-bridge.github.io
title EgoBridge: Domain Adaptation for Generalizable Imitation from Egocentric Human Data
topic Robotics
Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2509.19626