Saved in:
Bibliographic Details
Main Authors: Che, Yuchen, Wu, Jingtu, Zheng, Hao, Kanezaki, Asako
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.00493
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918361620807680
author Che, Yuchen
Wu, Jingtu
Zheng, Hao
Kanezaki, Asako
author_facet Che, Yuchen
Wu, Jingtu
Zheng, Hao
Kanezaki, Asako
contents Estimating the 6DoF pose of a novel object with a single reference view is challenging due to occlusions, view-point changes, and outliers. A core difficulty lies in finding robust cross-view correspondences, as existing methods often rely on discrete one-to-one matching that is non-differentiable and tends to collapse onto sparse key-points. We propose Confidence-aware Optimal Geometric Correspondence (COG), an unsupervised framework that formulates correspondence estimation as a confidence-aware optimal transport problem. COG produces balanced soft correspondences by predicting point-wise confidences and injecting them as optimal transport marginals, suppressing non-overlapping regions. Semantic priors from vision foundation models further regularize the correspondences, leading to stable pose estimation. This design integrates confidence into the correspondence finding and pose estimation pipeline, enabling unsupervised learning. Experiments show unsupervised COG achieves comparable performance to supervised methods, and supervised COG outperforms them.
format Preprint
id arxiv_https___arxiv_org_abs_2603_00493
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle COG: Confidence-aware Optimal Geometric Correspondence for Unsupervised Single-reference Novel Object Pose Estimation
Che, Yuchen
Wu, Jingtu
Zheng, Hao
Kanezaki, Asako
Computer Vision and Pattern Recognition
Estimating the 6DoF pose of a novel object with a single reference view is challenging due to occlusions, view-point changes, and outliers. A core difficulty lies in finding robust cross-view correspondences, as existing methods often rely on discrete one-to-one matching that is non-differentiable and tends to collapse onto sparse key-points. We propose Confidence-aware Optimal Geometric Correspondence (COG), an unsupervised framework that formulates correspondence estimation as a confidence-aware optimal transport problem. COG produces balanced soft correspondences by predicting point-wise confidences and injecting them as optimal transport marginals, suppressing non-overlapping regions. Semantic priors from vision foundation models further regularize the correspondences, leading to stable pose estimation. This design integrates confidence into the correspondence finding and pose estimation pipeline, enabling unsupervised learning. Experiments show unsupervised COG achieves comparable performance to supervised methods, and supervised COG outperforms them.
title COG: Confidence-aware Optimal Geometric Correspondence for Unsupervised Single-reference Novel Object Pose Estimation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2603.00493