Depth-Guided Monocular Object Pose Estimation for Warehouse Automation
Authors:
- Phan Xuan TanPTPhan Xuan Tan
- Dinh-Cuong HoangDHDinh-Cuong Hoang
- Anh-Nhat NguyenANAnh-Nhat Nguyen
- Eiji KamiokaEKEiji Kamioka
- Ta Huu Anh DuongTDTa Huu Anh Duong
- Tuan-Minh HuynhTHTuan-Minh Huynh
- Duc-Manh NguyenDNDuc-Manh Nguyen
- Duc-Huy NgoDNDuc-Huy Ngo
- Thu-Uyen NguyenTNThu-Uyen Nguyen
- Van-Thiep NguyenVNVan-Thiep Nguyen
- Duc-Thanh TranDTDuc-Thanh Tran
Open Access
- 25 June 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 110166-110184
Abstract
Accurate object pose estimation is essential for warehouse automation, enabling tasks such as robotic picking, sorting, and inventory management. Current state-of-the-art approaches rely on both color (RGB) and depth (D) images, as depth information provides critical geometric cues that enhance object localization and improve robustness against occlusions. However, RGBD-based methods require specialized depth sensors, which can be costly and may not function reliably in warehouse environments with reflective surfaces, varying lighting conditions, or sensor occlusions. To address these limitations, researchers have explored RGB-only approaches, but the absence of depth cues makes it challenging to handle occlusions, estimate object geometry, and differentiate between textureless or highly similar objects, which are common in warehouses. In this paper, we propose a novel end-to-end depth-guided object pose estimation method tailored for warehouse automation. Our approach leverages both depth and color images during training but relies solely on RGB images during inference. Depth images are used to supervise the training of a depth estimation network, which generates initial depth-aware features. These features are then refined using our proposed depth-guided feature enhancement module to improve spatial understanding and robustness. The enhanced features are subsequently utilized for keypoint-based 6D object pose estimation. By integrating depth-guided feature learning, our method significantly enhances pose estimation accuracy, especially in cluttered warehouse environments with severe occlusions and textureless objects. Extensive experiments on warehouse-specific datasets, as well as standard benchmark datasets, demonstrate that our approach outperforms existing RGB-based methods while maintaining real-time inference speeds, making it a highly practical solution for real-world warehouse automation applications.Keywords
This publication has 48 references indexed in Scilit:
- TPSFusion: A Transformer-based pyramid screening fusion network for 6D pose estimationImage and Vision Computing, 2025
- FormerPose: An efficient multi-scale fusion Transformer network based on RGB-D for 6D pose estimationJournal of Visual Communication and Image Representation, 2024
- Cross-modal attention and geometric contextual aggregation network for 6DoF object pose estimationNeurocomputing, 2024
- An integrated cost based approach for warehouse performance evaluation: A new multiphase modelAlexandria Engineering Journal, 2024
- A Transformer-based multi-modal fusion network for 6D pose estimationInformation Fusion, 2024
- Warehouse inventory management system using IoT and open source frameworkAlexandria Engineering Journal, 2018
- Implicit 3D Orientation Learning for 6D Object Detection from RGB Images2018
- Learning 6D Object Pose Estimation Using 3D Object Coordinates2014
- Model Based Training, Detection and Pose Estimation of Texture-Less 3D Objects in Heavily Cluttered Scenes2013
- EPnP: An Accurate O(n) Solution to the PnP ProblemInternational Journal of Computer Vision, 2008