{"title":"基于GPU的海量飞机CAD模型压缩与渲染","authors":"Tan Dunming, Zhao Gang, Yu Lu","doi":"10.1109/ICVRV.2012.8","DOIUrl":null,"url":null,"abstract":"The total size of massive aircraft CAD models is usually up to several GBs, which exceed not only the storage capacity of memory, but also the rendering ability of graphics card. In this paper, we present compression and rendering methods by exploring the up-to-date GPU techniques. To fit into the memory, vertex data are compressed from float to byte type with bounding box information and then decompressed with GPU. Index data are in short or byte type according to the vertex size, while normal data are deleted and generated by GPU while rendering. To render in real-time, vertex buffer object is exploited instead of traditional display list for efficiency and GPU occlusion query culls occluded parts to lower the rendering load. Furthermore, deliberately designed GPU shaders are applied to optimize the traditional rendering pipeline. The experiment results show by the GPU based methods, the compression rates get up to 5.3, massive CAD models such as the regional jet can be compressed within 178 MB and fit into memory of personal computers, and the rendering frame rates achieve up to 40 with cheap graphics card. It's proved that our method maximizes the GPU capabilities to accelerate the real-time rendering performance of massive aircraft CAD models.","PeriodicalId":421789,"journal":{"name":"2012 International Conference on Virtual Reality and Visualization","volume":"61 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2012-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"GPU Based Compression and Rendering of Massive Aircraft CAD Models\",\"authors\":\"Tan Dunming, Zhao Gang, Yu Lu\",\"doi\":\"10.1109/ICVRV.2012.8\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The total size of massive aircraft CAD models is usually up to several GBs, which exceed not only the storage capacity of memory, but also the rendering ability of graphics card. In this paper, we present compression and rendering methods by exploring the up-to-date GPU techniques. To fit into the memory, vertex data are compressed from float to byte type with bounding box information and then decompressed with GPU. Index data are in short or byte type according to the vertex size, while normal data are deleted and generated by GPU while rendering. To render in real-time, vertex buffer object is exploited instead of traditional display list for efficiency and GPU occlusion query culls occluded parts to lower the rendering load. Furthermore, deliberately designed GPU shaders are applied to optimize the traditional rendering pipeline. The experiment results show by the GPU based methods, the compression rates get up to 5.3, massive CAD models such as the regional jet can be compressed within 178 MB and fit into memory of personal computers, and the rendering frame rates achieve up to 40 with cheap graphics card. It's proved that our method maximizes the GPU capabilities to accelerate the real-time rendering performance of massive aircraft CAD models.\",\"PeriodicalId\":421789,\"journal\":{\"name\":\"2012 International Conference on Virtual Reality and Visualization\",\"volume\":\"61 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2012-09-14\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2012 International Conference on Virtual Reality and Visualization\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICVRV.2012.8\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2012 International Conference on Virtual Reality and Visualization","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICVRV.2012.8","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
GPU Based Compression and Rendering of Massive Aircraft CAD Models
The total size of massive aircraft CAD models is usually up to several GBs, which exceed not only the storage capacity of memory, but also the rendering ability of graphics card. In this paper, we present compression and rendering methods by exploring the up-to-date GPU techniques. To fit into the memory, vertex data are compressed from float to byte type with bounding box information and then decompressed with GPU. Index data are in short or byte type according to the vertex size, while normal data are deleted and generated by GPU while rendering. To render in real-time, vertex buffer object is exploited instead of traditional display list for efficiency and GPU occlusion query culls occluded parts to lower the rendering load. Furthermore, deliberately designed GPU shaders are applied to optimize the traditional rendering pipeline. The experiment results show by the GPU based methods, the compression rates get up to 5.3, massive CAD models such as the regional jet can be compressed within 178 MB and fit into memory of personal computers, and the rendering frame rates achieve up to 40 with cheap graphics card. It's proved that our method maximizes the GPU capabilities to accelerate the real-time rendering performance of massive aircraft CAD models.