{"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/robustness-analysis-of-pedestrian-detectors","title":"Robustness Analysis of Pedestrian Detectors for Surveillance","arxiv_id":"1807.04562","date":"2018-07-12","proceeding":null,"authors":["Yuming Fang","Guanqun Ding","Yuan Yuan","Weisi Lin","Haiwen Liu"],"abstract":"To obtain effective pedestrian detection results in surveillance video, there\nhave been many methods proposed to handle the problems from severe occlusion,\npose variation, clutter background, \\emph{etc}. Besides detection accuracy, a\nrobust surveillance video system should be stable to video quality degradation\nby network transmission, environment variation, etc. In this study, we conduct\nthe research on the robustness of pedestrian detection algorithms to video\nquality degradation. The main contribution of this work includes the following\nthree aspects. First, a large-scale Distorted Surveillance Video Data Set\n(DSurVD) is constructed from high-quality video sequences and their\ncorresponding distorted versions. Second, we design a method to evaluate\ndetection stability and a robustness measure called Robustness Quadrangle,\nwhich can be adopted to visualize detection accuracy of pedestrian detection\nalgorithms on high-quality video sequences and stability with video quality\ndegradation. Third, the robustness of seven existing pedestrian detection\nalgorithms is evaluated by the built DSurVD. Experimental results show that the\nrobustness can be further improved for existing pedestrian detection\nalgorithms. Additionally, we provide much in-depth discussion on how different\ndistortion types influence the performance of pedestrian detection algorithms,\nwhich is important to design effective pedestrian detection algorithms for\nsurveillance. The DSurVD data set can be download from BaiduYunDisk,\nhttps://pan.baidu.com/s/1I9Kqj8rmubOYu7bkBfkUpA, Password: lqmc","url_abs":"http://arxiv.org/abs/1807.04562v2","url_pdf":"http://arxiv.org/pdf/1807.04562v2.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":[{"paper_slug":"robustness-analysis-of-pedestrian-detectors","repo_url":"https://github.com/gqding/PedestrianIQA","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"pedestrian-detection","task_name":"Pedestrian Detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}