F. F. Doria, Felipe B. C. L. Lima, L. Real, Vinicius R. G. Oliveira, R. Y. Takimoto, A. K. Sato, Hossein R. Najafabadi, Fábio S. G. Tsuzuki, M. S. Tsuzuki
{"title":"足球比赛中摄像机位置的确定","authors":"F. F. Doria, Felipe B. C. L. Lima, L. Real, Vinicius R. G. Oliveira, R. Y. Takimoto, A. K. Sato, Hossein R. Najafabadi, Fábio S. G. Tsuzuki, M. S. Tsuzuki","doi":"10.1109/INDUSCON51756.2021.9529586","DOIUrl":null,"url":null,"abstract":"The automation of soccer video analysis is an area of recent interest, particularly the 3D reconstruction of soccer matches from broadcast video. One important task is the camera center determination. The camera center is calculated from a single soccer match image. It is assumed that a model of the soccer field with known dimensions is accessible. Two different approaches are presented: 1. Camera matrix determination; and 2. Homography matrix determination. It is shown that the homography matrix can be determined using mapped points and lines. Both methods are compared. To improve the numerical stability, a normalization procedure is used. Both methods with normalization are also compared. The results showed that the combination of points and lines with normalization improved the obtained results.","PeriodicalId":344476,"journal":{"name":"2021 14th IEEE International Conference on Industry Applications (INDUSCON)","volume":"116 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Determination of Camera Position in Soccer Games\",\"authors\":\"F. F. Doria, Felipe B. C. L. Lima, L. Real, Vinicius R. G. Oliveira, R. Y. Takimoto, A. K. Sato, Hossein R. Najafabadi, Fábio S. G. Tsuzuki, M. S. Tsuzuki\",\"doi\":\"10.1109/INDUSCON51756.2021.9529586\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The automation of soccer video analysis is an area of recent interest, particularly the 3D reconstruction of soccer matches from broadcast video. One important task is the camera center determination. The camera center is calculated from a single soccer match image. It is assumed that a model of the soccer field with known dimensions is accessible. Two different approaches are presented: 1. Camera matrix determination; and 2. Homography matrix determination. It is shown that the homography matrix can be determined using mapped points and lines. Both methods are compared. To improve the numerical stability, a normalization procedure is used. Both methods with normalization are also compared. The results showed that the combination of points and lines with normalization improved the obtained results.\",\"PeriodicalId\":344476,\"journal\":{\"name\":\"2021 14th IEEE International Conference on Industry Applications (INDUSCON)\",\"volume\":\"116 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2021-08-15\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2021 14th IEEE International Conference on Industry Applications (INDUSCON)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/INDUSCON51756.2021.9529586\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2021 14th IEEE International Conference on Industry Applications (INDUSCON)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/INDUSCON51756.2021.9529586","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
The automation of soccer video analysis is an area of recent interest, particularly the 3D reconstruction of soccer matches from broadcast video. One important task is the camera center determination. The camera center is calculated from a single soccer match image. It is assumed that a model of the soccer field with known dimensions is accessible. Two different approaches are presented: 1. Camera matrix determination; and 2. Homography matrix determination. It is shown that the homography matrix can be determined using mapped points and lines. Both methods are compared. To improve the numerical stability, a normalization procedure is used. Both methods with normalization are also compared. The results showed that the combination of points and lines with normalization improved the obtained results.