Transportation is one of the largest segments that can benefit from actionable insights derived from data captured by sensors. Between traffic, signaling systems, transportation systems, infrastructure, and transit, the opportunity for insights from these sensors to make transportation systems smarter is immense. Unfortunately, there are several reasons why these potential benefits have not yet materialized. Poor data quality, the lack of labels for the data, and the lack of high quality models that can convert the data into actionable insights are some of the biggest impediments to unlocking the value of the data. There is also need for platforms that allow for appropriate analysis from edge to cloud, which will accelerate the development and deployment of these models. The AI City Challenge Workshop at CVPR 2018 will specifically focus on problems such as

  • Estimating traffic flow characteristics, such as speed
  • Leveraging unsupervised approaches to detect anomalies caused by crashes, stalled vehicles, etc. This is the only way to get the humans in the loop pay attention to meaningful visual information
  • Multi-sensor tracking, and object re-identification in urban environments

We solicit original contributions in these and related areas where computer vision and specifically deep learning has shown promise in achieving large scale practical deployment that will help make cities smarter.

To accelerate the research and development of techniques that rely less on supervised approaches and more on transfer learning, unsupervised and semi-supervised approaches that go beyond bounding boxes, we are organizing a competition as part of the Challenge, with results presented at our CVPR 2018 workshop. Along with competition results, we are soliciting original contributions of research papers on computer vision for AI City, including but not limited to:

  • vehicle and road user detection and tracking
  • vehicle type classification
  • vehicle re-identification and matching
  • vehicle/pedestrian behavior analysis, activity recognition, speed estimation, congestion analysis, anomaly detection, and event summarization
  • 3D reconstruction and visualization for transportation detection
  • vision algorithms for autonomous driving car or assisted driving
  • smart transportation developments/integration with smart city


  • Monday, June 18, 2018
  • Salt Lake City, UT
  • Calvin L. Rampton Salt Palace Convention Center - Room 355B

09:00 – 09:25

Introduction and Challenge Summary

The AI City Challenge    

Dr. Milind Naphade, NVIDIA Corporation

09:25 – 09:50


Video Analytics in Traffic Management

John Garofolo, NIST

09:50 – 10:30

Oral Presentations Session I

Video Analytics in Smart Transportation for the AIC’18 Challenge (Team 18)

Ming-Ching Chang, Yi Wei, Nenghui Song, Siwei Lyu

Challenges in Large Scale Traffic Video Analysis (Team 79)

Weitao Feng, Deyi Ji, Yiru Wang, Shuorong Chang, Hansheng Ren, Weihao Gan

10:30 – 11:00

Coffee Break

11:00 – 12:20

Oral Presentations Session II

Graph@FIT Submission to the AI City Challenge 2018 (Team 6)

Jakub Sochor, Jakub Špaňhel, Roman Juránek, Petr Dobeš, Adam Herout

AIC2018 Report: Traffic Video Research (Team 12)

Tingyu Mao, Wei Zhang, Haoyu He, Yanjun Lin, Vinay Kale, Alexander Stein, Zoran Kostic

Visual Speed Estimation and Abnormality Detection (Team 39)

Panagiotis Giannakeris, Vagia Kaltsa, Konstantinos Avgerinakis, Alexia Briassouli, Stefanos Vrochidis, Ioannis Kompatsiaris

Traffic Flow Analysis with Multiple Adaptive Vehicle Detectors and Landmark-based Scanlines (Team 4)

Minh-Triet Tran, Tung Dinh-Duy, Thanh-Dat Truong, Vinh Ton-That, Thanh-Nhon Do, Quoc-An Luong, Thanh-An Nguyen, Vinh-Tiep Nguyen, Minh N. Do

12:20 – 13:20


13:20 – 15:20

Oral Presentations Session III

Single-camera and Inter-camera Vehicle Tracking and 3D Speed Estimation based on Fusion of Visual and Semantic Features (Team 48)

Zheng Tang, Gaoang Wang, Hao Xiao, Aotian Zheng, Jenq-Neng Hwang

Geometry-aware Traffic Flow Analysis by Detection and Tracking (Team 78)

Honghui Shi, Zhonghao Wang, Yang Zhang, Xinchao Wang, Thomas Huang

Vehicle Re-identification with the Space-Time Prior (Team 37)

Chih-Wei Wu, Chih-Ting Liu, Chen-En Jiang, Wei-Chih Tu, Shao-Yi Chien

Unsupervised Anomaly Detection for Traffic Based on Background Modeling (Team 63)

Jiayi Wei, Jianfei Zhao, Yanyun Zhao

Semi-Automatic 2D Solution for Vehicle Speed Estimation from Monocular Videos (Team 65)

Amit Kumar, Pirazh Khorramshahi, Wei-An Lin, Prithviraj Dhar, Jun-Cheng Chen, Rama Chellappa

Dual-Mode Vehicle Motion Pattern Learning for High Performance Road Traffic Anomaly Detection (Team 15)

Yan Xu, Xi Quyang, Yu Cheng, Shining Yu, Lin Xiong, Sugiri Pranata, Shengmei Shen, Junliang Xing

15:20 – 16:00

Panel on Video Analytics in Traffic Management

Terry Adams, Program Manager, I-ARPA

David Ness, City of Dubuque, Traffic Engineer

David Kuehn, FHWA, USDOT

Milind Naphade, NVIDIA Corporation

Anuj Sharma, Iowa State University

John Garofolo, NIST

16:00 – 16:10

DeepStream 2.0

DeepStream 2.0

Farzin Aghdasi, NVIDIA Corporation

16:10 – 16:30

Coffee Break

16:30 – 17:00

Posters and Demos

Vehicle Tracking and Speed Estimation from Traffic Videos (Team 26)

Shuai Hua, Manika Kapoor, David C. Anastasiu

Traffic Speed Estimation from Video Data (Team 40)

Tingting Huang

AIC2018 AImagelab (Team 41)

Pedro Antonio Marín-Reyes, Andrea Palazzi, Luca Bergamini, Simone Calderara, Javier Lorenzo-Navarro, Rita Cucchiara

Posters and Demos from All Oral Presentations

17:00 – 17:30

Awards Ceremony


Milind Naphade

NVIDIA Corporation

Rama Chellappa

University of Maryland, College Park

Jenq-Neng Hwang

University of Washington, Seattle

Ming-Ching Chang

University at Albany – SUNY

Ming-Yu Liu

NVIDIA Research

Siwei Lyu

University at Albany – SUNY

David Anastasiu

San Jose State University

Anuj Sharma

Iowa State University

Zeyu Gao

San Jose State University