Drone LAMS
收藏资源简介:
Drone LAMS是一个基于无人机的面部检测数据集,由东南大学创建,旨在解决无人机高空飞行时面部检测性能低下的问题。该数据集包含从261个视频中捕获的超过43,531个标注和4,001张具有不同俯仰或偏航角度的图像,覆盖范围从-90°到90°。Drone LAMS在检测性能上显著优于现有无人机面部检测数据集,特别是在大俯仰和偏航角度下。该数据集广泛应用于无人机面部检测研究,致力于提升在复杂环境下的检测效率和准确性。
Drone LAMS is a drone-based face detection dataset developed by Southeast University, aiming to address the issue of poor face detection performance when drones fly at high altitudes. This dataset contains over 43,531 annotations captured from 261 videos and 4,001 images with varying pitch or yaw angles, ranging from -90° to 90°. Drone LAMS significantly outperforms existing drone-based face detection datasets in terms of detection performance, especially under large pitch and yaw angles. This dataset is widely adopted in drone-based face detection research, with the goal of enhancing detection efficiency and accuracy in complex environments.

- 1Drone LAMS: A Drone-based Face Detection Dataset with Large Angles and Many Scenarios东南大学 · 2021年



