FADE
收藏资源简介:
FADE数据集是由武汉大学等机构创建的,专门用于建筑周围落物检测的视频数据集。该数据集包含1,881个视频,涵盖18个场景、8种落物类别、4种天气条件和4种视频分辨率。数据集的创建过程包括招募志愿者进行为期一年半的数据收集,使用高动态范围相机记录落物事件。FADE数据集的应用领域主要是公共安全,旨在通过智能视频监控技术有效检测和预防建筑周围落物事件,减少潜在的人身伤害风险。
The FADE dataset, developed by Wuhan University and other institutions, is a video dataset specifically designed for falling object detection around buildings. It comprises 1,881 videos covering 18 scenes, 8 falling object categories, 4 weather conditions, and 4 video resolutions. The dataset construction process involved recruiting volunteers for a 1.5-year data collection campaign, during which high dynamic range (HDR) cameras were utilized to record falling object incidents. The primary application of the FADE dataset lies in public safety, with the goal of effectively detecting and preventing falling object incidents around buildings via intelligent video surveillance technology to reduce the risk of potential personal injuries.

- 1FADE: A Dataset for Detecting Falling Objects around Buildings in Video武汉大学 · 2024年



