Solving Traffic4Cast Competition with U-Net and Temporal Domain Adaptation

TL;DR

This paper presents a U-Net based approach with domain adaptation techniques to predict future traffic states across different cities, addressing temporal domain shifts caused by COVID-19, and achieving third place in the Traffic4Cast 2021 challenge.

Contribution

The paper introduces a novel combination of U-Net architecture with pre-trained encoders and domain adaptation methods for traffic prediction under domain shift conditions.

Findings

Achieved third place in Traffic4Cast 2021 competition.

Effective use of domain adaptation techniques for temporal shift.

Demonstrated the viability of U-Net with pre-trained encoders for traffic forecasting.

Abstract

In this technical report, we present our solution to the Traffic4Cast 2021 Core Challenge, in which participants were asked to develop algorithms for predicting a traffic state 60 minutes ahead, based on the information from the previous hour, in 4 different cities. In contrast to the previously held competitions, this year's challenge focuses on the temporal domain shift in traffic due to the COVID-19 pandemic. Following the past success of U-Net, we utilize it for predicting future traffic maps. Additionally, we explore the usage of pre-trained encoders such as DenseNet and EfficientNet and employ multiple domain adaptation techniques to fight the domain shift. Our solution has ranked third in the final competition. The code is available at https://github.com/jbr-ai-labs/traffic4cast-2021.