OpenTie: Open-vocabulary Sequential Rebar Tying System

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Liu, Mingze, Fan, Sai, Li, Haozhen, Liang, Haobo, Yuan, Yixing, Wang, Yanke
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915470686289920
author Liu, Mingze
Fan, Sai
Li, Haozhen
Liang, Haobo
Yuan, Yixing
Wang, Yanke
author_facet Liu, Mingze
Fan, Sai
Li, Haozhen
Liang, Haobo
Yuan, Yixing
Wang, Yanke
contents Robotic practices on the construction site emerge as an attention-attracting manner owing to their capability of tackle complex challenges, especially in the rebar-involved scenarios. Most of existing products and research are mainly focused on flat rebar setting with model training demands. To fulfill this gap, we propose OpenTie, a 3D training-free rebar tying framework utilizing a RGB-to-point-cloud generation and an open-vocabulary detection. We implements the OpenTie via a robotic arm with a binocular camera and guarantees a high accuracy by applying the prompt-based object detection method on the image filtered by our propose post-processing procedure based a image to point cloud generation framework. The system is flexible for horizontal and vertical rebar tying tasks and the experiments on the real-world rebar setting verifies that the effectiveness of the system in practice.
format Preprint
id arxiv_https___arxiv_org_abs_2509_00064
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle OpenTie: Open-vocabulary Sequential Rebar Tying System
Liu, Mingze
Fan, Sai
Li, Haozhen
Liang, Haobo
Yuan, Yixing
Wang, Yanke
Robotics
Computer Vision and Pattern Recognition
Robotic practices on the construction site emerge as an attention-attracting manner owing to their capability of tackle complex challenges, especially in the rebar-involved scenarios. Most of existing products and research are mainly focused on flat rebar setting with model training demands. To fulfill this gap, we propose OpenTie, a 3D training-free rebar tying framework utilizing a RGB-to-point-cloud generation and an open-vocabulary detection. We implements the OpenTie via a robotic arm with a binocular camera and guarantees a high accuracy by applying the prompt-based object detection method on the image filtered by our propose post-processing procedure based a image to point cloud generation framework. The system is flexible for horizontal and vertical rebar tying tasks and the experiments on the real-world rebar setting verifies that the effectiveness of the system in practice.
title OpenTie: Open-vocabulary Sequential Rebar Tying System
topic Robotics
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.00064