Semantic2D
收藏资源简介:
Semantic2D是由天普大学创建的2D激光雷达语义分割数据集,旨在增强移动机器人在不同室内环境中的语义场景理解能力。该数据集包含在六个不同室内环境中收集的数据,涵盖九种典型室内物体类别。数据集的创建过程采用了半自动语义标注框架,通过手动标注环境地图和迭代最近点(ICP)算法,最小化人工标注工作。Semantic2D数据集的应用领域主要集中在移动机器人导航、多物体检测与跟踪、语义建图和自主导航等,旨在解决2D激光雷达在室内环境中的语义理解问题。
Semantic2D is a 2D LiDAR semantic segmentation dataset developed by Temple University, which is designed to enhance the semantic scene understanding capability of mobile robots in various indoor environments. This dataset comprises data collected across six distinct indoor environments, covering nine typical indoor object categories. The dataset was constructed using a semi-automatic semantic annotation framework, which minimizes manual annotation workload by utilizing manually annotated environmental maps and the Iterative Closest Point (ICP) algorithm. The primary application scenarios of the Semantic2D dataset focus on mobile robot navigation, multi-object detection and tracking, semantic mapping, autonomous navigation and other related fields, aiming to address the challenges of semantic understanding for 2D LiDAR in indoor environments.
Semantic2D: A Semantic Dataset for 2D Lidar Semantic Segmentation
概述
- 名称: Semantic2D
- 类型: 语义数据集
- 应用: 2D LiDAR 语义分割
更新信息
- 状态: 即将更新




