TIMo (Time-of-flight Indoor Monitoring)
收藏资源简介:
TIMo数据集由德国人工智能研究中心创建,专注于室内空间监控,使用最新一代的时间飞行相机捕捉红外和深度视频。数据集包含约1500个序列,用于异常检测和人员检测及计数,总计超过612,000个视频帧。创建过程中,数据在两个不同地点通过顶部或倾斜视角进行录制,并提供了详细的3D边界框和分割掩码注释。该数据集旨在推动室内空间监控算法的开发,特别是在保护隐私的前提下实现监控功能。
The TIMo dataset was created by the German Research Center for Artificial Intelligence (DFKI), focusing on indoor spatial surveillance. It captures infrared and depth videos using the latest generation time-of-flight (ToF) cameras. The dataset contains approximately 1,500 sequences designed for anomaly detection, human detection and counting, with a total of over 612,000 video frames. During its development, the data was recorded at two distinct locations from top-down or oblique viewing angles, and comes with detailed annotations including 3D bounding boxes and segmentation masks. This dataset aims to advance the development of indoor spatial surveillance algorithms, particularly to enable surveillance capabilities while safeguarding privacy.



