WiSARD
收藏资源简介:
WiSARD是由华盛顿大学航空航天系创建的一个大规模多模态视觉和热图像数据集,专为野外搜索和救援(WiSAR)任务设计。该数据集包含约56,000个标记图像,来源于无人机在不同地形、季节、天气和光照条件下的飞行。WiSARD是首个为自主WiSAR操作收集的大型多模态传感器数据集。数据集的创建旨在为研究人员提供一个多样化和挑战性的基准,以测试其算法在真实世界(生命救援)应用中的鲁棒性。WiSARD涵盖了多种环境,包括森林、田野、岩石区、海岸线和雪地,以及不同的光照和天气条件,为计算机视觉任务提供了丰富的挑战。
WiSARD is a large-scale multimodal visual and thermal imaging dataset developed by the Department of Aeronautics and Astronautics at the University of Washington, tailored specifically for wilderness search and rescue (WiSAR) missions. The dataset comprises approximately 56,000 labeled images, collected via drone flights across diverse terrains, seasons, weather conditions, and lighting environments. WiSARD is the first large-scale multimodal sensor dataset collected for autonomous WiSAR operations. The dataset was created to provide researchers with a diverse and challenging benchmark for testing the robustness of their algorithms in real-world life-saving rescue applications. WiSARD covers a wide range of environments including forests, fields, rocky areas, coastlines, and snow-covered landscapes, alongside varying lighting and weather conditions, offering rich challenges for computer vision tasks.




