UrbanFM: Inferring Fine-Grained Urban Flows

6 Feb 2019  ·  Yuxuan Liang, Kun Ouyang, Lin Jing, Sijie Ruan, Ye Liu, Junbo Zhang, David S. Rosenblum, Yu Zheng ·

Urban flow monitoring systems play important roles in smart city efforts around the world. However, the ubiquitous deployment of monitoring devices, such as CCTVs, induces a long-lasting and enormous cost for maintenance and operation. This suggests the need for a technology that can reduce the number of deployed devices, while preventing the degeneration of data accuracy and granularity. In this paper, we aim to infer the real-time and fine-grained crowd flows throughout a city based on coarse-grained observations. This task is challenging due to two reasons: the spatial correlations between coarse- and fine-grained urban flows, and the complexities of external impacts. To tackle these issues, we develop a method entitled UrbanFM based on deep neural networks. Our model consists of two major parts: 1) an inference network to generate fine-grained flow distributions from coarse-grained inputs by using a feature extraction module and a novel distributional upsampling module; 2) a general fusion subnet to further boost the performance by considering the influences of different external factors. Extensive experiments on two real-world datasets, namely TaxiBJ and HappyValley, validate the effectiveness and efficiency of our method compared to seven baselines, demonstrating the state-of-the-art performance of our approach on the fine-grained urban flow inference problem.

PDF Abstract

Datasets


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Fine-Grained Urban Flow Inference TaxiBJ-P1 UrbanFM MSE 15.6025 # 2
MAE 2.011 # 1
MAPE 0.327 # 1
Fine-Grained Urban Flow Inference TaxiBJ-P1 ESPCN MSE 17.6904 # 7
MAE 2.497 # 8
MAPE 0.732 # 8
Fine-Grained Urban Flow Inference TaxiBJ-P1 DeepSD MSE 17.2723 # 4
MAE 2.368 # 5
MAPE 0.614 # 5
Fine-Grained Urban Flow Inference TaxiBJ-P1 VDSR MSE 17.2972 # 5
MAE 2.213 # 3
MAPE 0.467 # 4
Fine-Grained Urban Flow Inference TaxiBJ-P1 SRResNet MSE 17.3388 # 6
MAE 2.457 # 6
MAPE 0.713 # 6
Fine-Grained Urban Flow Inference TaxiBJ-P1 UrbanFM-ne MSE 16.1202 # 3
MAE 2.047 # 2
MAPE 0.332 # 2
Fine-Grained Urban Flow Inference TaxiBJ-P1 HA MSE 22.4770 # 9
MAE 2.251 # 4
MAPE 0.336 # 3
Fine-Grained Urban Flow Inference TaxiBJ-P1 SRCNN MSE 18.4642 # 8
MAE 2.491 # 7
MAPE 0.714 # 7
Fine-Grained Urban Flow Inference TaxiBJ-P2 UrbanFM-ne MSE 19.2369 # 3
MAE 2.258 # 2
MAPE 0.320 # 2
Fine-Grained Urban Flow Inference TaxiBJ-P2 UrbanFM MSE 18.7402 # 2
MAE 2.224 # 1
MAPE 0.313 # 1
Fine-Grained Urban Flow Inference TaxiBJ-P3 UrbanFM MSE 20.2140 # 2
MAE 2.318 # 1
MAPE 0.315 # 1
Fine-Grained Urban Flow Inference TaxiBJ-P4 UrbanFM-ne MSE 12.666 # 3
MAE 1.845 # 2
MAPE 0.309 # 2
Fine-Grained Urban Flow Inference TaxiBJ-P4 UrbanFM MSE 12.2570 # 2
MAE 1.815 # 1
MAPE 0.308 # 1

Methods


No methods listed for this paper. Add relevant methods here