WHU数据集
收藏资源简介:
WHU数据集是由武汉大学遥感信息工程学院创建的,旨在支持大规模多视角立体重建任务。该数据集包含1776张虚拟航空图像,每张图像大小为5376×5376像素,覆盖面积约6.7×2.2平方公里。数据集通过精确的3D数字表面模型生成,包含完整的深度信息和相机参数,适用于训练和评估深度学习模型。该数据集特别适用于城市规模的三维重建,有助于解决大规模场景的立体匹配问题。
The WHU Dataset was created by the School of Remote Sensing and Information Engineering, Wuhan University, and is designed to support large-scale multi-view stereo reconstruction tasks. This dataset contains 1776 virtual aerial images, each with a resolution of 5376×5376 pixels, covering an area of approximately 6.7×2.2 square kilometers. It is generated from precise 3D digital surface models, and includes complete depth information and camera parameters, making it suitable for training and evaluating deep learning models. This dataset is particularly applicable for urban-scale 3D reconstruction, and helps address the stereo matching issues in large-scale scenes.

- 1A Novel Recurrent Encoder-Decoder Structure for Large-Scale Multi-view Stereo Reconstruction from An Open Aerial Dataset武汉大学遥感信息工程学院 · 2020年



