Abstract
Accurate citywide crowd activity prediction (CAP) can enable proactive crowd mobility management and timely responses to urban events, which has become increasingly important for a myriad of smart city planning and management purposes. However, complex correlations across the crowd activities, spatial and temporal urban environment features and their interactive dependencies, and relevant external factors (e.g., weather conditions) make it highly challenging to predict crowd activities accurately in terms of different venue categories (for instance, venues related to dining, services, and residence) and varying degrees (e.g., daytime and nighttime).
To address the above concerns, we propose
- [1] 2009–2010. Gowalla Data. Retrieved from https://go.gowalla.com/Google Scholar
- [2] 2012. Foursquare Data. Retrieved from https://foursquare.com/Google Scholar
- [3] 2021. Weather Data API. Retrieved from https://api.weather.com/Google Scholar
- [4] . 2021. Forecasting parking lots availability: Analysis from a real-world deployment. In IEEE PerCom Workshops. IEEE, 299–304.Google Scholar
- [5] . 2019. Understanding metropolitan crowd mobility via mobile cellular accessing data. ACM Trans. Spatial Algor. Syst. 5, 2 (2019), 1–18.Google ScholarDigital Library
- [6] . 2022. Multi-variate time series forecasting on variable subsets. In ACM SIGKDD.Google Scholar
- [7] . 2017. SCA-CNN: Spatial and channel-wise attention in convolutional networks for image captioning. In IEEE CVPR. 5659–5667.Google Scholar
- [8] . 2019. Automatic detection of urban features from wheelchair users’ movements. In IEEE PerCom. IEEE, 1–10.Google Scholar
- [9] . 2011. Performing a check-in: Emerging practices, norms and “conflicts” in location-sharing using Foursquare. In ACM MobileHCI. 57–66.Google Scholar
- [10] . 2021. ST-Norm: Spatial and temporal normalization for multi-variate time series forecasting. In ACM SIGKDD. 269–278.Google Scholar
- [11] . 2015. CityMomentum: An online approach for crowd behavior prediction at a citywide level. In ACM UbiComp. 559–569.Google Scholar
- [12] . 2018. Soft+ hardwired attention: An LSTM framework for human trajectory prediction and abnormal event detection. Neural Netw. 108 (2018), 466–478.Google ScholarCross Ref
- [13] . 2019. Crowd replication: Sensing-assisted quantification of human behavior in public spaces. ACM Trans. Spatial Algor. Syst. 5, 3 (2019), 1–34.Google ScholarDigital Library
- [14] . 2016. FCCF: Forecasting citywide crowd flows based on big data. In ACM SIGSPATIAL. 1–10.Google Scholar
- [15] . 2021. Cross-interaction hierarchical attention networks for urban anomaly prediction. In IJCAI. 4359–4365.Google Scholar
- [16] . 2021. Multi-head spatio-temporal attention mechanism for urban anomaly event prediction. Proc. ACMInteract., Mob., Wear. Ubiq. Technol. 5, 3 (2021), 1–21.Google ScholarDigital Library
- [17] . 2017. Estimating people flow from spatiotemporal population data via collective graphical mixture models. ACM Trans. Spatial Algor. Syst. 3, 1 (2017), 1–18.Google ScholarDigital Library
- [18] . 2019. Deepurbanevent: A system for predicting citywide crowd dynamics at big events. In ACM SIGKDD. 2114–2122.Google Scholar
- [19] . 2018. Geoman: Multi-level attention networks for geo-sensory time series prediction.. In IJCAI. 3428–3434.Google Scholar
- [20] . 2018. Attentive crowd flow machines. In ACM MM. 1553–1561.Google Scholar
- [21] . 2019. ADCrowdNet: An attention-injective deformable convolutional network for crowd understanding. In IEEE CVPR. 3225–3234.Google Scholar
- [22] . 2017. Temporal attention-gated model for robust sequence classification. In IEEE CVPR. 6730–6739.Google Scholar
- [23] . 2016. Personalized group recommender systems for location-and event-based social networks. ACM Trans. Spatial Algor. Syst. 2, 4 (2016), 1–29.Google ScholarDigital Library
- [24] . 2017. A dual-stage attention-based recurrent neural network for time series prediction. arXiv preprint arXiv:1704.02971 (2017).Google Scholar
- [25] . 2020. Activity recommendation: Optimizing life in the long term. In IEEE PerCom. IEEE, 1–10.Google Scholar
- [26] . 2018. What will you do for the rest of the day? An approach to continuous trajectory prediction. Proc. ACMInteract., Mob., Wear. Ubiq. Technol. 2, 4 (2018), 1–26.Google ScholarDigital Library
- [27] . 2015. Convolutional LSTM network: A machine learning approach for precipitation nowcasting. arXiv preprint arXiv:1506.04214 (2015).Google Scholar
- [28] . 2014. Prediction of human emergency behavior and their mobility following large-scale disaster. In ACM SIGKDD. 5–14.Google Scholar
- [29] . 2021. The impact of stationarity, regularity, and context on the predictability of individual human mobility. ACM Trans. Spatial Algor. Syst. 7, 4 (2021), 1–24.Google ScholarDigital Library
- [30] . 2018. Modeling and predicting bike demand in large city situations. In IEEE PerCom. IEEE, 1–10.Google Scholar
- [31] . 2019. Avoiding stress driving: Online trip recommendation from driving behavior prediction. In IEEE PerCom. IEEE, 1–10.Google Scholar
- [32] . 2017. Residual attention network for image classification. In IEEE CVPR. 3156–3164.Google Scholar
- [33] . 2019. Accurate detection of road network anomaly by understanding crowd’s driving strategies from human mobility. ACM Trans. Spatial Algor. Syst. 5, 2 (2019), 1–17.Google ScholarDigital Library
- [34] . 2020. Multi-task adversarial spatial-temporal networks for crowd flow prediction. In ACM CIKM. 1555–1564.Google Scholar
- [35] . 2020. A social force based pedestrian motion model considering multi-pedestrian interaction with a vehicle. ACM Trans. Spatial Algor. Syst. 6, 2 (2020), 1–27.Google ScholarDigital Library
- [36] . 2014. Modeling user activity preference by leveraging user spatial temporal characteristics in LBSNs. IEEE Trans. Syst., Man, Cybern.: Syst. 45, 1 (2014), 129–142.Google ScholarCross Ref
- [37] . 2021. Mobi-Track: Distilling direct path in time domain for high accuracy WiFi tracking. In IEEE PerCom Workshops. IEEE, 287–292.Google Scholar
- [38] . 2017. Deep spatio-temporal residual networks for citywide crowd flows prediction. In AAAI. AAAI Press, 1655–1661.Google Scholar
- [39] . 2019. GEVR: An event venue recommendation system for groups of mobile users. Proc. ACMInteract., Mob., Wear. Ubiq. Technol. 3, 1 (2019), 1–25.Google ScholarDigital Library
- [40] . 2018. Predicting multi-step citywide passenger demands using attention-based neural networks. In ACM WSDM. 736–744.Google Scholar
Index Terms
- STICAP: Spatio-temporal Interactive Attention for Citywide Crowd Activity Prediction
Recommendations
Deep spatio-temporal residual networks for citywide crowd flows prediction
AAAI'17: Proceedings of the Thirty-First AAAI Conference on Artificial IntelligenceForecasting the flow of crowds is of great importance to traffic management and public safety, and very challenging as it is affected by many complex factors, such as inter-region traffic, events, and weather. We propose a deep-learning-based approach, ...
Deep spatio-temporal neural network based on interactive attention for traffic flow prediction
AbstractTraffic flow forecasting is of great significance to urban traffic control and public safety applications. The key challenge of traffic flow forecasting is how to capture the complex correlation of different time levels and learn time dependence. ...
Spatio-temporal hard attention learning for skeleton-based activity recognition
Highlights- The novel problem of finding spatio-temporal hard attention in skeleton videos using deep learning for human activity recognition is discovered.
Graphical abstractDisplay Omitted
AbstractThe use of skeleton data for activity recognition has become prevalent due to its advantages over RGB data. A skeleton video includes frames showing two- or three-dimensional coordinates of human body joints. For recognizing an ...
Comments