Hideto Komori, Mariko Isogawa, Dan Mikami, Takasuke Nagai, Yoshimitsu Aoki
{"title":"用于人体运动分类的时间加权运动历史图像","authors":"Hideto Komori, Mariko Isogawa, Dan Mikami, Takasuke Nagai, Yoshimitsu Aoki","doi":"10.1007/s12283-023-00437-1","DOIUrl":null,"url":null,"abstract":"Abstract Vision-based human activity classification has remarkable potential for various applications in the sports context (e.g., motion analysis for performance enhancement, active sensing for athletes, etc.). Recently, learning-based human activity classifications have been widely researched. However, in sports scenes in which more detailed and player-specific classifications are required, this is a quite challenging task; in many cases, only a limited number of datasets are available, unlike daily movements such as walking or climbing stairs. Therefore, this paper proposes a time-weighted motion history image, an effective image sequence representation for learning-based human activity classification. Unlike conventional MHI based on the assumption that “the newer frame is more important,” our method generates importance-aware representation so that the predictor can “see” the frames that contribute to analyzing the specific human activity. Experimental results have shown the superiority of our method.","PeriodicalId":46387,"journal":{"name":"Sports Engineering","volume":"46 1","pages":"0"},"PeriodicalIF":1.5000,"publicationDate":"2023-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Time-weighted motion history image for human activity classification in sports\",\"authors\":\"Hideto Komori, Mariko Isogawa, Dan Mikami, Takasuke Nagai, Yoshimitsu Aoki\",\"doi\":\"10.1007/s12283-023-00437-1\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Abstract Vision-based human activity classification has remarkable potential for various applications in the sports context (e.g., motion analysis for performance enhancement, active sensing for athletes, etc.). Recently, learning-based human activity classifications have been widely researched. However, in sports scenes in which more detailed and player-specific classifications are required, this is a quite challenging task; in many cases, only a limited number of datasets are available, unlike daily movements such as walking or climbing stairs. Therefore, this paper proposes a time-weighted motion history image, an effective image sequence representation for learning-based human activity classification. Unlike conventional MHI based on the assumption that “the newer frame is more important,” our method generates importance-aware representation so that the predictor can “see” the frames that contribute to analyzing the specific human activity. Experimental results have shown the superiority of our method.\",\"PeriodicalId\":46387,\"journal\":{\"name\":\"Sports Engineering\",\"volume\":\"46 1\",\"pages\":\"0\"},\"PeriodicalIF\":1.5000,\"publicationDate\":\"2023-10-24\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Sports Engineering\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1007/s12283-023-00437-1\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"SPORT SCIENCES\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Sports Engineering","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1007/s12283-023-00437-1","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"SPORT SCIENCES","Score":null,"Total":0}
Time-weighted motion history image for human activity classification in sports
Abstract Vision-based human activity classification has remarkable potential for various applications in the sports context (e.g., motion analysis for performance enhancement, active sensing for athletes, etc.). Recently, learning-based human activity classifications have been widely researched. However, in sports scenes in which more detailed and player-specific classifications are required, this is a quite challenging task; in many cases, only a limited number of datasets are available, unlike daily movements such as walking or climbing stairs. Therefore, this paper proposes a time-weighted motion history image, an effective image sequence representation for learning-based human activity classification. Unlike conventional MHI based on the assumption that “the newer frame is more important,” our method generates importance-aware representation so that the predictor can “see” the frames that contribute to analyzing the specific human activity. Experimental results have shown the superiority of our method.
期刊介绍:
Sports Engineering is an international journal publishing original papers on the application of engineering and science to sport. The journal intends to fill the niche area which lies between classical engineering and sports science and aims to bridge the gap between the analysis of the equipment and of the athlete. Areas of interest include the mechanics and dynamics of sport, the analysis of movement, instrumentation, equipment design, surface interaction, materials and modelling. These topics may be applied to technology in almost any sport. The journal will be of particular interest to Engineering, Physics, Mathematics and Sports Science Departments and will act as a forum where research, industry and the sports sector can exchange knowledge and innovative ideas.