Laying the Foundations of Deep Long-Term Crowd Flow Prediction
Laying the Foundations of Deep Long-Term Crowd Flow Prediction
复制标题
DOI：
10.1007/978-3-030-58526-6_42
复制
发表时间：
2020-08
期刊：
影响因子：
--
通讯作者：
Samuel S. Sohn;H. Zhou;Seonghyeon Moon;Sejong Yoon;V. Pavlovic;M. Kapadia
中科院分区：
文献类型：
--
作者：
Samuel S. Sohn;H. Zhou;Seonghyeon Moon;Sejong Yoon;V. Pavlovic;M. Kapadia
预测复杂环境中的人群行为是人群和灾害管理、建筑设计和城市规划的关键要求。考虑到人群的即时状态，当前的方法必须在多个时间步上连续重复才能进行长期预测，从而导致计算成本昂贵且容易出错的结果。然而，大多数应用程序需要能够实时准确预测数百种可能的模拟结果（例如，在不同的环境和人群情况下），而这些方法的成本高昂。我们提出了第一个深度框架来即时预测任意大的现实环境中人群的长期流动。我们方法的核心是一种新颖的表征CAGE，它可以有效地将人群场景编码为紧凑的、固定大小的表征，无损地表征环境，以及用于即时长期人群流量预测的改进的SegNet架构。我们对新颖的合成数据集和真实数据集进行了全面的实验。我们的结果表明，我们的方法能够捕捉很长一段时间内真实人群运动的本质，同时推广到从未见过的环境和人群背景。相关的补充材料、模型和数据集可在 github.com/SSSohn/LTCF 上获取。
Predicting the crowd behavior in complex environments is a key requirement for crowd and disaster management, architectural design, and urban planning. Given a crowd’s immediate state, current approaches must be successively repeated over multiple time-steps for long-term predictions, leading to compute expensive and error-prone results. However, most applications require the ability to accurately predict hundreds of possible simulation outcomes (e.g., under different environment and crowd situations) at real-time rates, for which these approaches are prohibitively expensive. We propose the first deep framework to instantly predict thelong-term flowof crowds in arbitrarily large, realistic environments. Central to our approach are a novel representationCAGE, which efficiently encodes crowd scenarios into compact, fixed-size representations that losslessly represent the environment, and a modified SegNet architecture for instant long-term crowd flow prediction. We conduct comprehensive experiments on novel synthetic and real datasets. Our results indicate that our approach is able to capture the essence of real crowd movement over very long time periods, while generalizing to never-before-seen environments and crowd contexts. The associated Supplementary Material, models, and datasets are available at github.com/SSSohn/LTCF .