Utilizing UNet for the future traffic map prediction task Traffic4cast challenge 2020

2020/11/24 by Sungbin Choi, Choi, Sungbin
Computer Science · Engineering · #68T07 #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #FOS: Electrical engineering #Image and Video Processing (eess.IV) #Machine Learning (cs.LG) #cs.CV #cs.LG #eess.IV #electronic engineering #information engineering #msc:68T07

paper · pdf · doi:10.48550/arxiv.2012.00125

8 pages, 4 figures, NeurIPS 2020 Traffic4cast workshop paper

arxiv created 2020/11/24 · arxiv updated 2020/12/02

Abstract

This paper describes our UNet based experiments on the Traffic4cast challenge 2020. Similar to the Traffic4cast challenge 2019, the task is to predict traffic flow volume, direction and speed on a high resolution map of three large cities worldwide. We mainly experimented with UNet based deep convolutional networks with various compositions of densely connected convolution layers, average pooling layers and max pooling layers. Three base UNet model types are tried and predictions are combined by averaging prediction scores or taking median value. Our method achieved best performance in this years newly built challenge dataset.

Related

• Traffic map prediction using UNet based deep convolutional neural network
• Traffic4cast -- Large-scale Traffic Prediction using 3DResNet and Sparse-UNet
• Traffic4cast 2020 -- Graph Ensemble Net and the Importance of Feature And Loss Function Design for Traffic Prediction
• Traffic4cast at NeurIPS 2022 -- Predict Dynamics along Graph Edges from Sparse Node Data: Whole City Traffic and ETA from Stationary Vehicle Detectors
• SwinUNet3D -- A Hierarchical Architecture for Deep Traffic Prediction using Shifted Window Transformers