arXiv Robotics — research abstracts· Dhanushka Balasingham, Shamod Ginigaddarage, Hanojhan Rajahrajasingh, Nuwan Kodagoda, Rajitha de Silva·· 1 days agoEditorial score32
UltraLight Luma: Edge-Deployable Perception Network for Crop-Row Segmentation
UltraLight Luma: A Novel Edge-Deployable Perception Network for Crop-Row Segmentation in Agricultural Robotics
Summary
UltraLight Luma enables efficient crop-row detection with 13.1k parameters. It outperforms U-Net and YOLO variants while using 25.23 mJ per inference. Suitable for low-resource agricultural robots.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
Source：arXiv Robotics — research abstracts · arxiv.org