RipVIS
收藏资源简介:
RipVIS数据集是一个大规模、高质量的rip current实例分割数据集,由维尔茨堡大学计算机视觉实验室和布加勒斯特大学共同创建。该数据集包含来自全球不同地点的184个视频,总计212,328帧,其中150个视频含有rip current,其余34个视频不含rip current。这些视频展示了在不同视角、环境条件下的rip current,旨在为rip current的自动检测提供多样化、全面的训练和测试数据。
The RipVIS dataset is a large-scale, high-quality instance segmentation dataset focused on rip currents, co-created by the Computer Vision Lab at the University of Würzburg and the University of Bucharest. It consists of 184 videos collected from various locations across the globe, totaling 212,328 frames. Among these, 150 videos contain rip current instances, while the remaining 34 videos do not. These videos showcase rip current scenarios under diverse perspectives and environmental conditions, aiming to provide diversified, comprehensive training and testing data for the automatic detection of rip currents.

- 1RipVIS: Rip Currents Video Instance Segmentation Benchmark for Beach Monitoring and Safety德国维尔茨堡大学计算机视觉实验室 · 2025年



