Saved in:
Bibliographic Details
Main Authors: Zhang, Michael, Ying, Wei, Chen, Fangwen, Bai, Shifeng, Kang, Hanwen
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.02759
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910099765723136
author Zhang, Michael
Ying, Wei
Chen, Fangwen
Bai, Shifeng
Kang, Hanwen
author_facet Zhang, Michael
Ying, Wei
Chen, Fangwen
Bai, Shifeng
Kang, Hanwen
contents Accurate 6D object pose estimation is a fundamental capability for embodied agents, yet remains highly challenging in open-world environments. Many existing methods often rely on closed-set assumptions or geometry-agnostic regression schemes, limiting their generalization, stability, and real-time applicability in robotic systems. We present OMNI-PoseX, a vision foundation model that introduces a novel network architecture unifying open-vocabulary perception with an SO(3)-aware reflected flow matching pose predictor. The architecture decouples object-level understanding from geometry-consistent rotation inference, and employs a lightweight multi-modal fusion strategy that conditions rotation-sensitive geometric features on compact semantic embeddings, enabling efficient and stable 6D pose estimation. To enhance robustness and generalization, the model is trained on large-scale 6D pose datasets, leveraging broad object diversity, viewpoint variation, and scene complexity to build a scalable open-world pose backbone. Comprehensive evaluations across benchmark pose estimation, ablation studies, zero-shot generalization, and system-level robotic grasping integration demonstrate the effectiveness of OMNI-PoseX. The OMNI-PoseX achieves SOTA pose accuracy and real-time efficiency, while delivering geometrically consistent predictions that enable reliable grasping of diverse, previously unseen objects.
format Preprint
id arxiv_https___arxiv_org_abs_2604_02759
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle OMNI-PoseX: A Fast Vision Model for 6D Object Pose Estimation in Embodied Tasks
Zhang, Michael
Ying, Wei
Chen, Fangwen
Bai, Shifeng
Kang, Hanwen
Robotics
Accurate 6D object pose estimation is a fundamental capability for embodied agents, yet remains highly challenging in open-world environments. Many existing methods often rely on closed-set assumptions or geometry-agnostic regression schemes, limiting their generalization, stability, and real-time applicability in robotic systems. We present OMNI-PoseX, a vision foundation model that introduces a novel network architecture unifying open-vocabulary perception with an SO(3)-aware reflected flow matching pose predictor. The architecture decouples object-level understanding from geometry-consistent rotation inference, and employs a lightweight multi-modal fusion strategy that conditions rotation-sensitive geometric features on compact semantic embeddings, enabling efficient and stable 6D pose estimation. To enhance robustness and generalization, the model is trained on large-scale 6D pose datasets, leveraging broad object diversity, viewpoint variation, and scene complexity to build a scalable open-world pose backbone. Comprehensive evaluations across benchmark pose estimation, ablation studies, zero-shot generalization, and system-level robotic grasping integration demonstrate the effectiveness of OMNI-PoseX. The OMNI-PoseX achieves SOTA pose accuracy and real-time efficiency, while delivering geometrically consistent predictions that enable reliable grasping of diverse, previously unseen objects.
title OMNI-PoseX: A Fast Vision Model for 6D Object Pose Estimation in Embodied Tasks
topic Robotics
url https://arxiv.org/abs/2604.02759