The 5th AI City Challenge

@article{Naphade2021The5A,
  title={The 5th AI City Challenge},
  author={Milind R. Naphade and Shuo Wang and D. Anastasiu and Zheng Tang and Ming-Ching Chang and Xiaodong Yang and Yue Yao and Liang Zheng and Pranamesh Chakraborty and Christian E. L{\'o}pez and Anuj Sharma and Qi Feng and Vitaly Ablavsky and Stan Sclaroff},
  journal={2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)},
  year={2021},
  pages={4258-4268}
}
  • M. Naphade, Shuo Wang, S. Sclaroff
  • Published 25 April 2021
  • Computer Science
  • 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
The AI City Challenge was created with two goals in mind: (1) pushing the boundaries of research and development in intelligent video analysis for smarter cities use cases, and (2) assessing tasks where the level of performance is enough to cause real-world adoption. Transportation is a segment ripe for such adoption. The fifth AI City Challenge attracted 305 participating teams across 38 countries, who leveraged city-scale real traffic data and high-quality synthetic data to compete in five… 

Figures and Tables from this paper

The 4th AI City Challenge
TLDR
The 4th annual edition of the AI City Challenge has attracted 315 participating teams, who leverage city-scale real traffic data and high-quality synthetic data to compete in four challenge tracks, and results show promise that AI technology can enable smarter and safer transportation systems.
Scalable and Real-time Multi-Camera Vehicle Detection, Re-Identification, and Tracking
TLDR
This work proposes a real-time city-scale multi-camera vehicle tracking system that compares favorably to computationally intensive alternatives and handles real-world, low-resolution CCTV instead of idealized and curated video streams.
TrackNet: A Triplet metric-based method for Multi-Target Multi-Camera Vehicle Tracking
TLDR
This work presents TrackNet, a method for Multi-Target Multi-Camera (MTMC) vehicle tracking from traffic video sequences, based on a modular approach that tracks detections through single camera using Kalman filter, and matches tracks by a triplet metric learning strategy.
Rope3D: TheRoadside Perception Dataset for Autonomous Driving and Monocular 3D Object Detection Task
TLDR
The first high-diversity challenging roadside perception 3D dataset- Rope3D from a novel view is presented and the existing frontal-view monocular 3D object detection approaches are tailor to solve the inherent ambi-guities caused by various sensors, viewpoints.
Vehicle Detection and Tracking for 511 Traffic Cameras With U-Shaped Dual Attention Inception Neural Networks and Spatial-Temporal Map
TLDR
The U-shaped dual attention inception (DAIU) deep-learning model was designed, given the similarities between the STMap vehicle detection task and the medical image segmentation task, and has shown better performance than other mainstream neural networks based on segmentation model evaluation metrics.
New Directions in Traffic Control Analysis through Video Surveillance
TLDR
An eco system is developed for object detection, vehicle number detection and the speed of the vehicle using computer vision algorithms and the application tool developed can warn the speedy drivers on the fly.
Robust 2D/3D Vehicle Parsing in Arbitrary Camera Views for CVIS
TLDR
This work presents a novel approach to robustly detect and perceive vehicles in different camera views as part of a cooperative vehicle-infrastructure system (CVIS), and proposes a part-assisted novel view synthesis algorithm for data augmentation to deal with multi-view data scarcity.
Deep Learning-Based Object Tracking via Compressed Domain Residual Frames
TLDR
This work uses exclusively the residual frames already generated by the video compression codec to train and test the network and shows that using residual frames for deep learning-based object tracking can be just as effective as using classical decoded frames.

References

SHOWING 1-10 OF 65 REFERENCES
The 4th AI City Challenge
TLDR
The 4th annual edition of the AI City Challenge has attracted 315 participating teams, who leverage city-scale real traffic data and high-quality synthetic data to compete in four challenge tracks, and results show promise that AI technology can enable smarter and safer transportation systems.
The 2019 AI City Challenge
TLDR
Participation in this challenge has grown five-fold this year as tasks have become more relevant to traffic optimization and challenging to the computer vision community.
The 2018 NVIDIA AI City Challenge
TLDR
The second edition of the NVIDIA AI City Challenge provided a forum to more than 70 academic and industrial research teams to compete and solve real-world problems using traffic camera video data.
The NVIDIA AI City Challenge
  • M. Naphade, D. Anastasiu, J. Gao
  • Computer Science
    2017 IEEE SmartWorld, Ubiquitous Intelligence & Computing, Advanced & Trusted Computed, Scalable Computing & Communications, Cloud & Big Data Computing, Internet of People and Smart City Innovation (SmartWorld/SCALCOM/UIC/ATC/CBDCom/IOP/SCI)
  • 2017
TLDR
The NVIDIA AI City Challenge brought together 29 teams from universities in 4 continents to collaboratively annotate a 125 hour data set and then compete on detection, localization and classification tasks as well as traffic and safety application analytics tasks.
Connecting Language and Vision for Natural Language-Based Vehicle Retrieval
TLDR
This work proposes to jointly train the state-of-the-art vision models with the transformer-based language model in an end-to-end manner to connect language and vision, and achieves the 1st place on the 5th AI City Challenge.
An Empirical Study of Vehicle Re-Identification on the AI City Challenge
  • Haowen Luo, Weihua Chen, Hao Li
  • Computer Science
    2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
  • 2021
TLDR
The solution for the Track2 in AI City Challenge 2021 (AICITY21) is introduced, a vehicle re-identification (ReID) task with both the real-world data and synthetic data, and unsupervised domain-adaptive (UDA) training, post-processing, model ensembling.
A Strong Baseline for Vehicle Re-Identification
TLDR
This paper analyzes the main factors hindering the Vehicle Re-ID performance, and presents solutions, specifically targeting the dataset Track 2 of the 5th AI City Challenge, including reducing the domain gap between real and synthetic data, and adaptive loss weight adjustment.
A Robust MTMC Tracking System for AI-City Challenge 2021
  • Jinxing Ye, Xipeng Yang, Xiao Tan
  • Computer Science
    2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
  • 2021
TLDR
This paper proposes a practical framework for dealing with the MTMC problem, which associates all candidate trajectories between two successive cameras using the established distance matrix, and combines all successively matching results for final submission.
Practices and A Strong Baseline for Traffic Anomaly Detection
TLDR
A straightforward and efficient framework that includes preprocessing, a dynamic track module, and post-processing for traffic anomaly detection is proposed that was ranked 1 in the NVIDIA AI CITY 2021 leaderboard.
CityFlow: A City-Scale Benchmark for Multi-Target Multi-Camera Vehicle Tracking and Re-Identification
TLDR
This work introduces CityFlow, a city-scale traffic camera dataset consisting of more than 3 hours of synchronized HD videos from 40 cameras across 10 intersections, with the longest distance between two simultaneous cameras being 2.5 km.
...
...