• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to project page

2020 Fiscal Year Research-status Report

A Benchmark for Video-Like Urban Computing on Citywide Crowd and Traffic Prediction

Research Project

Project/Area Number 20K19859
Research InstitutionThe University of Tokyo

Principal Investigator

姜 仁河  東京大学, 情報基盤センター, 助教 (20865266)

Project Period (FY) 2020-04-01 – 2022-03-31
Keywordssmart city / society 5.0 / crowd density / crowd flow / human mobility
Outline of Annual Research Achievements

Predicting the density and flow of the crowd or traffic at a citywide level becomes possible by using the big data and cutting-edge AI technologies. It has been a very significant research topic with high social impact, which can be widely applied to emergency management, traffic regulation, and urban planning.

In this year, we have the following achivements: (1) We propose a new concept called video-like urban computing (4D tensor prediction), based on which, a series of research problems could be uniformly defined and formulated. (2) We publish a new dataset for crowd density and in-out flow prediction together with Yahoo Japan Research, which is generated based on a real-world smartphone app. (3) We also collect multiple open datasets on crowd density, taxi demand, traffic accident.

Current Status of Research Progress
Current Status of Research Progress

2: Research has progressed on the whole more than it was originally planned.

Reason

Originally, we have set the following four targets.
(1) Collect the open datasets for crowd density, crowd flow, and taxi demand. (2) Try to publish new datasets for crowd density and crowd flow that have higher quality than the existing ones. (3) Try to implement the state-of-the-art deep learning models for crowd density prediction, crowd flow prediction, and taxi demand prediction. (4) Based on the above, develop a dominant model for a series of mesh-based prediction problems.

In the past year, we have finished (1)(2)(3) and part of (4). In the next year, we will continue to develop a superior model for crowd density and flow prediction. Moreover, we would like to apply our model to other similar tasks such as eletricity demand prediction, air quality prediction, and so on.

Strategy for Future Research Activity

(1) We will explore the limitations of the state-of-the-arts and try to integrate the advanced techniques from different models to build a new and super-dominant model.

(2) We will extend those models and algorithms to be applicable to other related research problems other than citywide crowd/traffic prediction, such as, citywide air quality prediction, citywide power consumption prediction, citywide transportation demand prediction and so on.

Causes of Carryover

Due to the COVID-19, I could not attend many international conferences, therefore, I could not use the travel budgets as planned.

The COVID-19 situation is still lasting this year, so I think the travel budgets for this year will also be left. Instead, I try to move the travel budgets to the product budgets to buy a new GPU server by using around 2 millon yens.

  • Research Products

    (3 results)

All 2021 2020

All Journal Article (2 results) Presentation (1 results)

  • [Journal Article] Transfer Urban Human Mobility via POI Embedding over Multiple Cities2021

    • Author(s)
      Jiang Renhe、Song Xuan、Fan Zipei、Xia Tianqi、Wang Zhaonan、Chen Quanjun、Cai Zekun、Shibasaki Ryosuke
    • Journal Title

      ACM/IMS Transactions on Data Science

      Volume: 2 Pages: 1~26

    • DOI

      10.1145/3416914

  • [Journal Article] DeepCrowd: A Deep Model for Large-Scale Citywide Crowd Density and Flow Prediction2021

    • Author(s)
      Jiang Renhe、Cai Zekun、Wang Zhaonan、Yang Chuang、Fan Zipei、Chen Quanjun、Tsubouchi Kota、Song Xuan、Shibasaki Ryosuke
    • Journal Title

      IEEE Transactions on Knowledge and Data Engineering

      Volume: 0 Pages: 1~1

    • DOI

      10.1109/TKDE.2021.3077056

  • [Presentation] DualSIN: Dual Sequential Interaction Network for Human Intentional Mobility Prediction2020

    • Author(s)
      Chen Quanjun、Jiang Renhe、Yang Chuang、Cai Zekun、Fan Zipei、Tsubouchi Kota、Shibasaki Ryosuke、Song Xuan
    • Organizer
      SIGSPATIAL '20: Proceedings of the 28th International Conference on Advances in Geographic Information Systems

URL: 

Published: 2021-12-27  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi