由于视频序列和异常事件都是由连续的片段组成的，因此这些片段之间存在时间依赖关系。因此，同时学习视频中相邻片段之间的局部时间依赖关系和相邻片段之间的全局时间依赖关系，有利于提高模型的检测性能。然而，以前的弱监督方法忽略了片段之间的长期和短期时间依赖关系。为了缓解上述问题，本文通过引入片段之间的时间依赖关系，提出了一种端到端局部-全局时间依赖网络，使模型更加关注片段之间的关联。翻译成英文

As both video sequences and anomalous events are composed of consecutive segments, there exists a temporal dependency between these segments. Therefore, simultaneously learning the local temporal dependency between adjacent segments within the video and the global temporal dependency between adjacent segments can improve the detection performance of the model. However, previous weakly supervised methods neglected the long-term and short-term temporal dependency between segments. To address this issue, this paper proposes an end-to-end local-global temporal dependency network by introducing the temporal dependency between segments, which enables the model to focus more on the correlations between segments.