arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1905.03578cs.CVcs.AIcs.ITcs.ROmath.IT
收藏

预测未来活动的学习表征

Learning Representations for Predicting Future Activities

University of Freiburg(弗莱堡大学);University of Toronto(多伦多大学);Peking University(北京大学)

Mohammadreza Zolfaghari, Özgün Çiçek, Syed Mohsin Ali, Farzaneh Mahdisoltani, Can Zhang, Thomas Brox

摘要:

预见未来是智力的关键因素之一。它涉及对过去和当前环境的理解,以及对其可能的动态的适当经验。在这项工作中,我们解决了未来的预测在抽象层次的活动。我们提出了一个网络模块,用于以自我监督的方式学习环境动态的嵌入。考虑到未来活动的模糊性和高方差,我们使用了一个多假设方案,可以代表多个未来。我们通过对Epic-Kitterfly和Breakfast数据集上的未来活动进行分类来演示该方法。此外,我们生成描述未来活动的标题

英文摘要:

Foreseeing the future is one of the key factors of intelligence. It involves understanding of the past and current environment as well as decent experience of its possible dynamics. In this work, we address future prediction at the abstract level of activities. We propose a network module for learning embeddings of the environment's dynamics in a self-supervised way. To take the ambiguities and high variances in the future activities into account, we use a multi-hypotheses scheme that can represent multiple futures. We demonstrate the approach by classifying future activities on the Epic-Kitchens and Breakfast datasets. Moreover, we generate captions that describe the future activities

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑