{"title":"基于图像过渡网络的手部形状估计","authors":"Y. Hamada, N. Shimada, Y. Shirai","doi":"10.1109/HUMO.2000.897387","DOIUrl":null,"url":null,"abstract":"We present a method of hand posture estimation from silhouette images taken by two cameras. First, we extract the silhouette contour for a pair of images. We construct an eigenspace from images of hands with various postures. For effective matching, we define a shape complexity for each image to see how well the shape feature is represented. For a pair of input images, the total matching error is computed by combining the two matching errors according to the shape complexity. Thus the best-matched image is obtained for a pair of images. For rapid processing, we limit the matching candidate by using the constraint on the shape change. The possible shape transition is represented by a transition network. Because the network is hard to build, we apply offline learning, where nodes and links are automatically created by showing examples of hand shape sequences. We show experiments of building the transition networks and the performance of matching using the network.","PeriodicalId":384462,"journal":{"name":"Proceedings Workshop on Human Motion","volume":"46 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2000-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"13","resultStr":"{\"title\":\"Hand shape estimation using image transition network\",\"authors\":\"Y. Hamada, N. Shimada, Y. Shirai\",\"doi\":\"10.1109/HUMO.2000.897387\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"We present a method of hand posture estimation from silhouette images taken by two cameras. First, we extract the silhouette contour for a pair of images. We construct an eigenspace from images of hands with various postures. For effective matching, we define a shape complexity for each image to see how well the shape feature is represented. For a pair of input images, the total matching error is computed by combining the two matching errors according to the shape complexity. Thus the best-matched image is obtained for a pair of images. For rapid processing, we limit the matching candidate by using the constraint on the shape change. The possible shape transition is represented by a transition network. Because the network is hard to build, we apply offline learning, where nodes and links are automatically created by showing examples of hand shape sequences. We show experiments of building the transition networks and the performance of matching using the network.\",\"PeriodicalId\":384462,\"journal\":{\"name\":\"Proceedings Workshop on Human Motion\",\"volume\":\"46 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2000-12-07\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"13\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Proceedings Workshop on Human Motion\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/HUMO.2000.897387\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings Workshop on Human Motion","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/HUMO.2000.897387","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Hand shape estimation using image transition network
We present a method of hand posture estimation from silhouette images taken by two cameras. First, we extract the silhouette contour for a pair of images. We construct an eigenspace from images of hands with various postures. For effective matching, we define a shape complexity for each image to see how well the shape feature is represented. For a pair of input images, the total matching error is computed by combining the two matching errors according to the shape complexity. Thus the best-matched image is obtained for a pair of images. For rapid processing, we limit the matching candidate by using the constraint on the shape change. The possible shape transition is represented by a transition network. Because the network is hard to build, we apply offline learning, where nodes and links are automatically created by showing examples of hand shape sequences. We show experiments of building the transition networks and the performance of matching using the network.