FROG
收藏资源简介:
FROG数据集是一个专为2D激光雷达人体检测设计的大型数据集,由西班牙塞维利亚的Universidad Pablo de Olavide的服务机器人实验室创建。该数据集包含超过40万条激光扫描记录,每条记录均进行了100%的标注,相较于现有数据集如DROW,提供了更高的激光分辨率和扫描频率。数据集的场景设置在塞维利亚的皇家阿尔卡萨尔,一个每年接待超过150万游客的历史遗址,因此数据集涵盖了室内外多种复杂环境,包括人群密集区域和各种挑战性特征。FROG数据集不仅用于评估和比较2D激光雷达人体检测器,还支持深度学习模型的训练和测试,旨在解决机器人导航中的人体检测问题,提高机器人在复杂环境中的感知能力和交互性能。
The FROG dataset is a large-scale dataset specifically designed for 2D LiDAR-based human detection, created by the Service Robotics Laboratory at Universidad Pablo de Olavide in Seville, Spain. This dataset contains over 400,000 laser scan records, with each record being 100% annotated. Compared to existing datasets such as DROW, it provides higher laser resolution and scanning frequency. The dataset was collected at the Real Alcázar of Seville, a historic site that receives more than 1.5 million visitors annually. Consequently, it covers a wide range of complex indoor and outdoor environments, including densely crowded areas and various challenging features. The FROG dataset not only can be used to evaluate and compare 2D LiDAR-based human detectors, but also supports the training and testing of deep learning models. It aims to solve the problem of human detection in robotic navigation and improve the perceptual capabilities and interactive performance of robots in complex environments.

- 1FROG: A new people detection dataset for knee-high 2D range finders服务机器人实验室 · 2023年



