English

CAPT: Category-level Articulation Estimation from a Single Point Cloud Using Transformer

Computer Vision and Pattern Recognition 2024-02-28 v1 Artificial Intelligence Robotics

Abstract

The ability to estimate joint parameters is essential for various applications in robotics and computer vision. In this paper, we propose CAPT: category-level articulation estimation from a point cloud using Transformer. CAPT uses an end-to-end transformer-based architecture for joint parameter and state estimation of articulated objects from a single point cloud. The proposed CAPT methods accurately estimate joint parameters and states for various articulated objects with high precision and robustness. The paper also introduces a motion loss approach, which improves articulation estimation performance by emphasizing the dynamic features of articulated objects. Additionally, the paper presents a double voting strategy to provide the framework with coarse-to-fine parameter estimation. Experimental results on several category datasets demonstrate that our methods outperform existing alternatives for articulation estimation. Our research provides a promising solution for applying Transformer-based architectures in articulated object analysis.

Keywords

Cite

@article{arxiv.2402.17360,
  title  = {CAPT: Category-level Articulation Estimation from a Single Point Cloud Using Transformer},
  author = {Lian Fu and Ryoichi Ishikawa and Yoshihiro Sato and Takeshi Oishi},
  journal= {arXiv preprint arXiv:2402.17360},
  year   = {2024}
}

Comments

Accepted to ICRA 2024

R2 v1 2026-06-28T15:01:42.234Z