Self-Supervised Text-Driven Point Cloud Upsampling via Semantic Text Guidance
Journal of Imaging · 11.05.2026 · 12 (5) · 204
DOI: 10.3390/jimaging12050204 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Imaging, online erschienen am 11.05.2026. Verfasst von Zhiyong Zhang und 5 weiteren Autor:innen. Beteiligte Einrichtung: School of Computer Science and Engineering, Tianjin University of Technology, Tianjin 300000, China. Die Arbeit nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Point cloud upsampling is a fundamental task in 3D vision, yet most existing methods adopt a global and uniform strategy, which is computationally inefficient and fails to address the need for region-specific refinement. To address this challenge, we propose PartSPUNet, a novel self-supervised, text-driven point cloud upsampling framework designed to enhance robotic perception through task-oriented local refinement. Inspired by the human cognitive process where high-level language instructions guide visual attention to specific regions of interest, our method allows an operator to use intuitive natural language prompts to direct the upsampling process. Specifically, PartSPUNet leverages a pretrained vision–language model to zero-shot localize the user-specified semantic part within a sparse point cloud. It then performs geometry-aware densification exclusively on this target region, recovering rich geometric details while preserving the global structure. Experimental results demonstrate that our approach significantly outperforms existing methods in reconstructing specified areas, offering a powerful and intuitive tool for enhancing the 3D perception pipeline in intelligent robotic systems.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Computer Science and Engineering, Tianjin University of Technology, Tianjin 300000, China.
- School of Information Science and Technology, Hangzhou Normal University, Hangzhou 310000, China.
- Department of Technology, Management and Economics, Technical University of Denmark, 2800 Lyngby, Denmark.
ORCID-Profile: Zhiyong Zhang, Ruyu Liu, Jianhua Zhang
Zitieren
APA
Zhang, Z., Qiu, M., Chen, S., Liu, R., Zhang, J., & Chen, S. (2026). Self-Supervised Text-Driven Point Cloud Upsampling via Semantic Text Guidance. Journal of Imaging, 12(5), 204. https://doi.org/10.3390/jimaging12050204
Vancouver
Zhang Z, Qiu M, Chen S, Liu R, Zhang J, Chen S. Self-Supervised Text-Driven Point Cloud Upsampling via Semantic Text Guidance. Journal of Imaging. 2026;12(5):204. doi:10.3390/jimaging12050204
BibTeX
@article{Zhang2026,
author = {Zhang, Zhiyong and Qiu, Meiling and Chen, Shuo and Liu, Ruyu and Zhang, Jianhua and Chen, Shengyong},
title = {Self-Supervised Text-Driven Point Cloud Upsampling via Semantic Text Guidance},
journal = {Journal of Imaging},
year = {2026},
volume = {12},
number = {5},
pages = {204},
doi = {10.3390/jimaging12050204}
}
EUVIMED