Structure-aware completion of plant 3D LiDAR point clouds via a multi-resolution GAN-inversion network
These results indicate that MRC-Net advances unsupervised point cloud completion by balancing global shape consistency with fine-grained detail. The method provides a reliable data foundation for downstream tasks-including autonomous navigation, high-precision 3D modeling, and agricultural robotics-thereby contributing to improved data quality in precision-agriculture and related domains.
Authors
- Xinyu Xue
- Zhikai Zhang (ORCID: https://orcid.org/0000-0001-5978-9868)
- Zhihong Zhang (ORCID: https://orcid.org/0000-0002-3749-4716)
- Jingxin Shen
- Zhenyu Ma (ORCID: https://orcid.org/0000-0002-7984-1827)
- Yitian Sun
- Zhiming Wei (ORCID: https://orcid.org/0000-0002-9215-7149)
- Wu Liu (ORCID: https://orcid.org/0000-0001-5809-9785)
- Qinglong Li (ORCID: https://orcid.org/0000-0003-3782-6052)
- Xiaoju Li (ORCID: https://orcid.org/0000-0002-3887-6273)
- Jianing Long
Institutions
- Beijing University of Posts and Telecommunications (CN)
- Qingdao Agricultural University (CN)
- Shandong Academy of Agricultural Machinery Sciences (CN)
- China Telecom (China) (CN)
- Ministry of Agriculture and Rural Affairs (CN)
- Nanjing Institute of Agricultural Mechanization (CN)
- Shanghai Institute of Technology (CN)
Publication Details
- Journal
- Frontiers in Plant Science
- Published
- 2025-12-19
- DOI
- https://doi.org/10.3389/fpls.2025.1698843
- Primary Topic
- 3D Shape Modeling and Analysis
- Type
- article
- Field-Weighted Citation Impact
- 0.00