A Graph-based U-Net Model for Predicting Traffic in unseen Cities
This addresses traffic management for urban planners and drivers, but it is incremental as it builds on existing U-Net methods.
The paper tackled traffic prediction by combining U-Net with graph layers to improve spatial generalization to unseen cities, achieving better performance than a Vanilla U-Net.
Accurate traffic prediction is a key ingredient to enable traffic management like rerouting cars to reduce road congestion or regulating traffic via dynamic speed limits to maintain a steady flow. A way to represent traffic data is in the form of temporally changing heatmaps visualizing attributes of traffic, such as speed and volume. In recent works, U-Net models have shown SOTA performance on traffic forecasting from heatmaps. We propose to combine the U-Net architecture with graph layers which improves spatial generalization to unseen road networks compared to a Vanilla U-Net. In particular, we specialize existing graph operations to be sensitive to geographical topology and generalize pooling and upsampling operations to be applicable to graphs.