Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Hsiao, Tsu-Ching, Chen, Hao-Wei, Yang, Hsuan-Kung, Lee, Chun-Yi
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2305.15873
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866910400620003328
author Hsiao, Tsu-Ching
Chen, Hao-Wei
Yang, Hsuan-Kung
Lee, Chun-Yi
author_facet Hsiao, Tsu-Ching
Chen, Hao-Wei
Yang, Hsuan-Kung
Lee, Chun-Yi
contents Addressing pose ambiguity in 6D object pose estimation from single RGB images presents a significant challenge, particularly due to object symmetries or occlusions. In response, we introduce a novel score-based diffusion method applied to the $SE(3)$ group, marking the first application of diffusion models to $SE(3)$ within the image domain, specifically tailored for pose estimation tasks. Extensive evaluations demonstrate the method's efficacy in handling pose ambiguity, mitigating perspective-induced ambiguity, and showcasing the robustness of our surrogate Stein score formulation on $SE(3)$. This formulation not only improves the convergence of denoising process but also enhances computational efficiency. Thus, we pioneer a promising strategy for 6D object pose estimation.
format Preprint
id arxiv_https___arxiv_org_abs_2305_15873
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Confronting Ambiguity in 6D Object Pose Estimation via Score-Based Diffusion on SE(3)
Hsiao, Tsu-Ching
Chen, Hao-Wei
Yang, Hsuan-Kung
Lee, Chun-Yi
Computer Vision and Pattern Recognition
Addressing pose ambiguity in 6D object pose estimation from single RGB images presents a significant challenge, particularly due to object symmetries or occlusions. In response, we introduce a novel score-based diffusion method applied to the $SE(3)$ group, marking the first application of diffusion models to $SE(3)$ within the image domain, specifically tailored for pose estimation tasks. Extensive evaluations demonstrate the method's efficacy in handling pose ambiguity, mitigating perspective-induced ambiguity, and showcasing the robustness of our surrogate Stein score formulation on $SE(3)$. This formulation not only improves the convergence of denoising process but also enhances computational efficiency. Thus, we pioneer a promising strategy for 6D object pose estimation.
title Confronting Ambiguity in 6D Object Pose Estimation via Score-Based Diffusion on SE(3)
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2305.15873