EUVIMEDPoznaj EUVIMED
Open access · CC BY 4.0

Real-Time Visual Perception and Grasping in Unstructured Cluttered Environments using 3D Point Cloud Segmentation

Seo-Yun Cho, Alice Johnson

Computer Science Bulletin · 2026-01-30 · 9 (1) · 36-43

DOI: 10.71465/csb198 ↗

At a glance

Journal article in Computer Science Bulletin, published online on 2026-01-30. Written by Seo-Yun Cho and 1 co-authors. The full text is freely available (open access, CC BY 4.0).

AbstraktDane bibliograficzneCiteRelated studies

Abstract (original language)

The automation of robotic grasping in unstructured and cluttered environments remains a pivotal challenge in modern robotics and industrial automation. While structured bin-picking is relatively solved, scenarios involving overlapping objects, varying geometries, and chaotic arrangements present significant difficulties for traditional vision systems. This paper proposes a comprehensive framework for real-time visual perception and grasping, leveraging advanced 3D point cloudsegmentation techniques. We introduce a hybrid processing pipeline that integrates geometric filtering with deep learning-based semantic segmentation to isolate individual object instances within a dense clutter. By utilizing depth information directly from 3D point clouds rather than relying solely on 2D RGB data, the system achieves superior spatial understanding and occlusion management. The proposed method employs a region-growing segmentation algorithm enhanced by local surface normal estimation to generate precise grasp candidates. Furthermore, we implement a ranking mechanism that evaluates grasp stability based on antipodal geometry and collision risks. Experimental validation is conducted using a 6-DOF robotic arm equipped with a parallel-jaw gripper. The results demonstrate that our approach significantly outperforms baseline Euclidean clustering methods in terms of grasp success rate and recognition accuracy, while maintaining a computational latency suitable for real-time industrial applications.

Cite

APA

Cho, S., & Johnson, A. (2026). Real-Time Visual Perception and Grasping in Unstructured Cluttered Environments using 3D Point Cloud Segmentation. Computer Science Bulletin, 9(1), 36-43. https://doi.org/10.71465/csb198

Vancouver

Cho S, Johnson A. Real-Time Visual Perception and Grasping in Unstructured Cluttered Environments using 3D Point Cloud Segmentation. Computer Science Bulletin. 2026;9(1):36-43. doi:10.71465/csb198

BibTeX
@article{Cho2026,
  author = {Cho, Seo-Yun and Johnson, Alice},
  title = {Real-Time Visual Perception and Grasping in Unstructured Cluttered Environments using 3D Point Cloud Segmentation},
  journal = {Computer Science Bulletin},
  year = {2026},
  volume = {9},
  number = {1},
  pages = {36--43},
  doi = {10.71465/csb198}
}
Start for free →