📄 中文摘要
农业领域的动态和异质性给物体检测和定位带来了重大挑战，尤其是对于负责勘测之前未见的非结构化环境的自主移动机器人。同时，实时检测系统的需求日益增长，而这些系统不依赖于大规模手动标注的真实世界数据集。本研究提出了一种全面的注释到检测框架，旨在利用有限和部分标注的训练数据训练出稳健的多模态检测器。该方法结合了跨模态注释转移和早期传感器融合管道，并通过多阶段检测架构有效提升了检测性能。
一种用于移动农业机器人在田间自主且稳健的葡萄树干定位的注释到检测框架
📄 English Summary
An Annotation-to-Detection Framework for Autonomous and Robust Vine Trunk Localization in the Field by Mobile Agricultural Robots
The dynamic and heterogeneous nature of agricultural fields poses significant challenges for object detection and localization, particularly for autonomous mobile robots tasked with surveying previously unseen unstructured environments. There is an increasing demand for real-time detection systems that do not rely on large-scale manually labeled real-world datasets. This research presents a comprehensive annotation-to-detection framework designed to train a robust multi-modal detector using limited and partially labeled training data. The proposed methodology incorporates cross-modal annotation transfer and an early-stage sensor fusion pipeline, which, in conjunction with a multi-stage detection architecture, effectively enhances detection performance.