Dual Encoding U-Net for Spatio-Temporal Domain Shift Frame Prediction

The landscape of city-wide mobility behaviour has altered significantly over the past 18 months. The ability to make accurate and reliable predictions on such behaviour has likewise changed drastically with COVID-19 measures impacting how populations across the world interact with the different facets of mobility. This raises the question: "How does one use an abundance of pre-covid mobility data to make predictions on future behaviour in a present/post-covid environment?" This paper seeks to address this question by introducing an approach for traffic frame prediction using a lightweight Dual-Encoding U-Net built using only 12 Convolutional layers that incorporates a novel approach to skip-connections between Convolutional LSTM layers. This approach combined with an intuitive handling of training data can model both a temporal and spatio-temporal domain shift (gitlab.com/alchera/alchera-traffic4cast-2021).

[1]  Rado Gazo,et al.  TLab: Traffic Map Movie Forecasting Based on HR-NET , 2020, ArXiv.

[2]  Yoshua Bengio,et al.  Why Does Unsupervised Pre-training Help Deep Learning? , 2010, AISTATS.

[3]  Sungbin Choi,et al.  Utilizing UNet for the future traffic map prediction task Traffic4cast challenge 2020 , 2020, ArXiv.

[4]  Peter Bloem,et al.  Uncertainty Intervals for Graph-based Spatio-Temporal Traffic Prediction , 2020, ArXiv.

[5]  Qi Qi,et al.  Traffic4cast 2020 - Graph Ensemble Net and the Importance of Feature And Loss Function Design for Traffic Prediction , 2020, ArXiv.

[6]  Jonathan Tompson,et al.  Efficient object localization using Convolutional Networks , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[7]  Yunbo Wang,et al.  Towards Good Practices of U-Net for Traffic Forecasting , 2020, ArXiv.

[8]  Bin Han,et al.  Research on Real-Time Local Rainfall Prediction Based on MEMS Sensors , 2018, J. Sensors.

[9]  Pedro Herruzo,et al.  Recurrent Autoencoder with Skip Connections and Exogenous Variables for Traffic Forecasting , 2019, NeurIPS.

[10]  James Demmel,et al.  Large Batch Optimization for Deep Learning: Training BERT in 76 minutes , 2019, ICLR.

[11]  David P. Kreil,et al.  Traffic4cast at NeurIPS 2020 ? yet more on theunreasonable effectiveness of gridded geo-spatial processes , 2020, NeurIPS.

[12]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[13]  James T. Kwok,et al.  Generalizing from a Few Examples , 2019, ACM Comput. Surv..

[14]  Raia Hadsell,et al.  The surprising efficiency of framing geo-spatial time series forecasting as a video prediction task - Insights from the IARAI Traffic4cast Competition at NeurIPS 2019 , 2020, NeurIPS.

[15]  Pankaj Daga,et al.  Temporal Autoencoder with U-Net Style Skip-Connections for Frame Prediction , 2020, ArXiv.

[16]  Dit-Yan Yeung,et al.  Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting , 2015, NIPS.

[17]  Ye Hong,et al.  Traffic4cast-Traffic Map Movie Forecasting - Team MIE-Lab , 2019, ArXiv.