Smart City Gnosys

Smart city article details

Title Smart City Traffic Intersection: Impact Of Video Quality And Scene Complexity On Precision And Inference
ID_Doc 50553
Authors Duan Z.; Yang Z.; Samoilenko R.; Oza D.S.; Jagadeesan A.; Sun M.; Ye H.; Xiong Z.; Zussman G.; Kostic Z.
Year 2022
Published 2021 IEEE 23rd International Conference on High Performance Computing and Communications, 7th International Conference on Data Science and Systems, 19th International Conference on Smart City and 7th International Conference on Dependability in Sensor, Cloud and Big Data Systems and Applications, HPCC-DSS-SmartCity-DependSys 2021
DOI http://dx.doi.org/10.1109/HPCC-DSS-SmartCity-DependSys53884.2021.00226
Abstract Traffic intersections are prime locations for deployment of infrastructure sensors and edge computing nodes to realize the vision of a smart city. It is expected that the needs of a smart city, in regards to traffic and pedestrian traffic systems monitored by cameras/video, can be met by using state-of-the-art artificial-intelligence (AI) based object detectors and trackers. A critical component in designing an effective real-time object detection/tracking pipeline is the understanding of how object density, i.e., the number of objects in a scene, and image-resolution and frame rate influence the performance metrics. This study explores the accuracy and speed metrics with the goal of supporting pipelines that meet the precision and latency needs of a real-time environment. We examine the impact of varying image-resolution, frame rate and object-density on the object detection performance metrics. The experiments on the COSMOS testbed dataset show that varying the frame width from 416 pixels to 832 pixels, and cropping the images to a square resolution, result in the increase in average precision for all object classes. Decreasing the frame rate from 15 fps to 5 fps preserves more than 90% of the highest F1 score achieved for all object classes. The results inform the choice of video preprocessing stages, modifications to established AI-based object detection/tracking methods, and suggest optimal hyper-parameter values. © 2021 IEEE.
Author Keywords Deep Learning Models; Object Detection; Smart City; Video Resolution


Similar Articles

Id Similarity Authors Title Published
16287 View0.924Yang S.; Bailey E.; Yang Z.; Ostrometzky J.; Zussman G.; Seskar I.; Kostic Z.Cosmos Smart Intersection: Edge Compute And Communications For Bird'S Eye Object Tracking2020 IEEE International Conference on Pervasive Computing and Communications Workshops, PerCom Workshops 2020 (2020)
44316 View0.903Ghasemi M.; Kleisarchaki S.; Calmant T.; Gürgen L.; Ghaderi J.; Kostic Z.; Zussman G.Real-Time Camera Analytics For Enhancing Traffic Intersection SafetyMobiSys 2022 - Proceedings of the 2022 20th Annual International Conference on Mobile Systems, Applications and Services (2022)
44405 View0.9Ghasemi M.; Kleisarchaki S.; Calmant T.; Lu J.; Ojha S.; Kostic Z.; Gurgen L.; Zussman G.; Ghaderi J.Real-Time Multi-Camera Analytics For Traffic Information Extraction And Visualization2023 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2023 (2023)
5789 View0.894Ranka S.; Rangarajan A.; Elefteriadou L.; Srinivasan S.; Poasadas E.; Hoffman D.; Ponnulari R.; Dilmore J.; Byron T.A Vision Of Smart Traffic Infrastructure For Traditional, Connected, And Autonomous VehiclesProceedings - 2020 International Conference on Connected and Autonomous Driving, MetroCAD 2020 (2020)
35765 View0.893Fleck T.; Pavlitska S.; Nitzsche S.; Pachideh B.; Peccia F.; Ahmed S.H.; Meyer S.M.; Richter M.; Broertjes K.; Neftci E.; Becker J.; Bringmann O.; Marius Zollner J.Low-Power Traffic Surveillance Using Multiple Rgb And Event Cameras: A SurveyProceedings of 2023 IEEE International Smart Cities Conference, ISC2 2023 (2023)
41647 View0.891Tian J.; Jin Q.; Wang Y.; Yang J.; Zhang S.; Sun D.Performance Analysis Of Deep Learning-Based Object Detection Algorithms On Coco Benchmark: A Comparative StudyJournal of Engineering and Applied Science, 71, 1 (2024)
22093 View0.886Hua S.; Anastasiu D.C.Effective Vehicle Tracking Algorithm For Smart Traffic NetworksProceedings - 13th IEEE International Conference on Service-Oriented System Engineering, SOSE 2019, 10th International Workshop on Joint Cloud Computing, JCC 2019 and 2019 IEEE International Workshop on Cloud Computing in Robotic Systems, CCRS 2019 (2019)
21807 View0.885Skadins A.; Ivanovs M.; Rava R.; Nesenbergs K.Edge Pre-Processing Of Traffic Surveillance Video For Bandwidth And Privacy Optimization In Smart CitiesProceedings of the Biennial Baltic Electronics Conference, BEC, 2020-October (2020)
43514 View0.885Tubesing J.; Poole H.; Yu L.P.; Tsai Y.-C.J.Proposing A Comprehensive Evaluation Method For Ai-Based Traffic Detection System And Post-Processing Method Using Physical ConstraintsInternational Conference on Transportation and Development 2022: Application of Emerging Technologies - Selected Papers from the Proceedings of the International Conference on Transportation and Development 2022, 1 (2022)
17768 View0.885Liu K.Deep Associated Elastic Tracker For Intelligent Traffic IntersectionsAIChallengeIoT 2020 - Proceedings of the 2020 2nd International Workshop on Challenges in Artificial Intelligence and Machine Learning for Internet of Things (2020)