The 6th AI City Challenge

Milind Naphade,Shuo Wang,David C. Anastasiu,Zheng Tang,Ming-Ching Chang,Yue Yao,Liang Zheng,Mohammed Shaiqur Rahman,Archana Venkatachalapathy,Anuj Sharma,Qi Feng,Vitaly Ablavsky,Stan Sclaroff,Pranamesh Chakraborty,Alice Li,Shangru Li,Rama Chellappa

DOI: https://doi.org/10.48550/arXiv.2204.10380

2022-06-10

Abstract:The 6th edition of the AI City Challenge specifically focuses on problems in two domains where there is tremendous unlocked potential at the intersection of computer vision and artificial intelligence: Intelligent Traffic Systems (ITS), and brick and mortar retail businesses. The four challenge tracks of the 2022 AI City Challenge received participation requests from 254 teams across 27 countries. Track 1 addressed city-scale multi-target multi-camera (MTMC) vehicle tracking. Track 2 addressed natural-language-based vehicle track retrieval. Track 3 was a brand new track for naturalistic driving analysis, where the data were captured by several cameras mounted inside the vehicle focusing on driver safety, and the task was to classify driver actions. Track 4 was another new track aiming to achieve retail store automated checkout using only a single view camera. We released two leader boards for submissions based on different methods, including a public leader board for the contest, where no use of external data is allowed, and a general leader board for all submitted results. The top performance of participating teams established strong baselines and even outperformed the state-of-the-art in the proposed challenge tracks.

Computer Vision and Pattern Recognition

What problem does this paper attempt to address?

The problems that this paper attempts to solve mainly focus on two fields, namely Intelligent Traffic Systems (ITS) and physical retail business, unlocking great potential through the combination of computer vision and artificial intelligence technology. Specifically, the four challenge tracks in the AI City Challenge in 2022 mentioned in the paper focus on the following problems respectively:

1. **Urban - scale Multi - Target Multi - Camera (MTMC) Vehicle Tracking**: This track aims to develop technologies that can track vehicles in the city across multiple cameras, helping traffic engineers understand the travel time of the entire corridor, thereby optimizing traffic flow management and reducing congestion.

2. **Vehicle Trajectory Retrieval Based on Natural - Language Descriptions**: The task of this track is to retrieve specific vehicle trajectories according to the natural - language descriptions in the video, aiming to improve the efficiency and accuracy in scenarios such as traffic accident investigations and vehicle tracking.

3. **Natural Driving Behavior Analysis**: This track focuses on classifying drivers' behaviors through the data captured by cameras installed inside the vehicle, especially identifying behaviors that may lead to distracted driving, such as texting, making phone calls, yawning, etc., in order to enhance driving safety.

4. **Retail Store Automated Checkout**: The goal of this track is to use a single - view camera to achieve automatic identification and counting of retail products, reducing the waiting time for customers at checkout and providing a more seamless shopping experience.

Each track has designed specific tasks and evaluation indicators to assess the effectiveness and innovativeness of the technical solutions submitted by the participating teams. Through these challenges, researchers hope to promote technological progress in related fields and solve key problems in practical applications.