{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/cityflow-a-city-scale-benchmark-for-multi","title":"CityFlow: A City-Scale Benchmark for Multi-Target Multi-Camera Vehicle Tracking and Re-Identification","arxiv_id":"1903.09254","date":"2019-03-21","proceeding":"CVPR 2019 6","authors":["Zheng Tang","Milind Naphade","Ming-Yu Liu","Xiaodong Yang","Stan Birchfield","Shuo Wang","Ratnesh Kumar","David Anastasiu","Jenq-Neng Hwang"],"abstract":"Urban traffic optimization using traffic cameras as sensors is driving the\nneed to advance state-of-the-art multi-target multi-camera (MTMC) tracking.\nThis work introduces CityFlow, a city-scale traffic camera dataset consisting\nof more than 3 hours of synchronized HD videos from 40 cameras across 10\nintersections, with the longest distance between two simultaneous cameras being\n2.5 km. To the best of our knowledge, CityFlow is the largest-scale dataset in\nterms of spatial coverage and the number of cameras/videos in an urban\nenvironment. The dataset contains more than 200K annotated bounding boxes\ncovering a wide range of scenes, viewing angles, vehicle models, and urban\ntraffic flow conditions. Camera geometry and calibration information are\nprovided to aid spatio-temporal analysis. In addition, a subset of the\nbenchmark is made available for the task of image-based vehicle\nre-identification (ReID). We conducted an extensive experimental evaluation of\nbaselines/state-of-the-art approaches in MTMC tracking, multi-target\nsingle-camera (MTSC) tracking, object detection, and image-based ReID on this\ndataset, analyzing the impact of different network architectures, loss\nfunctions, spatio-temporal models and their combinations on task effectiveness.\nAn evaluation server is launched with the release of our benchmark at the 2019\nAI City Challenge (https://www.aicitychallenge.org/) that allows researchers to\ncompare the performance of their newest techniques. We expect this dataset to\ncatalyze research in this field, propel the state-of-the-art forward, and lead\nto deployed traffic optimization(s) in the real world.","url_abs":"http://arxiv.org/abs/1903.09254v4","url_pdf":"http://arxiv.org/pdf/1903.09254v4.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[],"tasks":[{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"vehicle-re-identification","task_name":"Vehicle Re-Identification"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[],"datasets_introduced":[{"slug":"cityflow","name":"CityFlow","full_name":""}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1903.09254","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}