Wuhan data and code

This dataset is collected from a residential area in Wuhan (China), which covers an area of 0.53km², including about 20 million faces. We generated the mesh data from oblique aerial images with a ground sample distance (GSD) of 5cm. We manually annotated the dataset and divided it into 7 object categories, namely <i>roof</i>, <i>façade</i>, <i>window</i>, <i>impervious surface</i>, <i>tree</i>, <i>vehicle</i> and <i>low vegetation</i>.The python code for texture-point-integrated network also included

本数据集采集自中国武汉某居民区，覆盖范围达0.53平方千米，内含约2000万个网格面元。我们采用地面采样距离（GSD）为5cm的倾斜航空影像生成了本数据集的网格数据。本数据集经人工标注，划分为7个目标类别，分别为屋顶（roof）、立面（façade）、窗户（window）、不透水面（impervious surface）、树木（tree）、车辆（vehicle）以及低矮植被（low vegetation）。本数据集亦包含纹理点融合网络（texture-point-integrated network）的Python代码。