Demo-Pose: Depth-Monocular Modality Fusion For Object Pose Estimation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Agarwal, Rachit, Joshi, Abhishek, Chalasani, Sathish, Kim, Woo Jin
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914429884432384
author Agarwal, Rachit
Joshi, Abhishek
Chalasani, Sathish
Kim, Woo Jin
author_facet Agarwal, Rachit
Joshi, Abhishek
Chalasani, Sathish
Kim, Woo Jin
contents Object pose estimation is a fundamental task in 3D vision with applications in robotics, AR/VR, and scene understanding. We address the challenge of category-level 9-DoF pose estimation (6D pose + 3Dsize) from RGB-D input, without relying on CAD models during inference. Existing depth-only methods achieve strong results but ignore semantic cues from RGB, while many RGB-D fusion models underperform due to suboptimal cross-modal fusion that fails to align semantic RGB cues with 3D geometric representations. We propose DeMo-Pose, a hybrid architecture that fuses monocular semantic features with depth-based graph convolutional representations via a novel multimodal fusion strategy. To further improve geometric reasoning, we introduce a novel Mesh-Point Loss (MPL) that leverages mesh structure during training without adding inference overhead. Our approach achieves real-time inference and significantly improves over state-of-the-art methods across object categories, outperforming the strong GPV-Pose baseline by 3.2\% on 3D IoU and 11.1\% on pose accuracy on the REAL275 benchmark. The results highlight the effectiveness of depth-RGB fusion and geometry-aware learning, enabling robust category-level 3D pose estimation for real-world applications.
format Preprint
id arxiv_https___arxiv_org_abs_2603_27533
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Demo-Pose: Depth-Monocular Modality Fusion For Object Pose Estimation
Agarwal, Rachit
Joshi, Abhishek
Chalasani, Sathish
Kim, Woo Jin
Computer Vision and Pattern Recognition
Artificial Intelligence
I.2.10
Object pose estimation is a fundamental task in 3D vision with applications in robotics, AR/VR, and scene understanding. We address the challenge of category-level 9-DoF pose estimation (6D pose + 3Dsize) from RGB-D input, without relying on CAD models during inference. Existing depth-only methods achieve strong results but ignore semantic cues from RGB, while many RGB-D fusion models underperform due to suboptimal cross-modal fusion that fails to align semantic RGB cues with 3D geometric representations. We propose DeMo-Pose, a hybrid architecture that fuses monocular semantic features with depth-based graph convolutional representations via a novel multimodal fusion strategy. To further improve geometric reasoning, we introduce a novel Mesh-Point Loss (MPL) that leverages mesh structure during training without adding inference overhead. Our approach achieves real-time inference and significantly improves over state-of-the-art methods across object categories, outperforming the strong GPV-Pose baseline by 3.2\% on 3D IoU and 11.1\% on pose accuracy on the REAL275 benchmark. The results highlight the effectiveness of depth-RGB fusion and geometry-aware learning, enabling robust category-level 3D pose estimation for real-world applications.
title Demo-Pose: Depth-Monocular Modality Fusion For Object Pose Estimation
topic Computer Vision and Pattern Recognition
Artificial Intelligence
I.2.10
url https://arxiv.org/abs/2603.27533