UAVStereo
收藏资源简介:
UAVStereo是由中国人民解放军战略支援部队信息工程大学创建的一个多分辨率立体匹配数据集,专注于无人机低空场景。该数据集包含超过34,000对立体图像,覆盖三种典型场景,旨在解决无人机图像在立体匹配中的应用问题。数据集通过结合无人机图像和LiDAR点云重建的详细网格,生成精确密集的视差图。UAVStereo不仅包含合成和真实立体对,还提供多分辨率和多场景图像对,以适应各种传感器和环境。此数据集的创建过程涉及复杂的数据采集和处理流程,旨在为无人机场景下的立体匹配研究提供高质量的数据支持。
UAVStereo is a multi-resolution stereo matching dataset developed by the Information Engineering University of the Strategic Support Force of the Chinese People's Liberation Army, focusing on low-altitude UAV scenarios. This dataset contains over 34,000 pairs of stereo images covering three typical scenarios, aiming to address the application challenges of UAV images in stereo matching. It generates accurate dense disparity maps by combining UAV images with detailed grids reconstructed from LiDAR point clouds. UAVStereo includes both synthetic and real stereo pairs, as well as multi-resolution and multi-scenario image pairs to adapt to various sensors and environments. The creation of this dataset involves complex data acquisition and processing workflows, with the goal of providing high-quality data support for stereo matching research in UAV scenarios.




