{"title":"虚拟视图合成质量改进","authors":"Tzu-Chin Lee, Chun-Liang Chien, H. Hang","doi":"10.1109/3DTV.2016.7548958","DOIUrl":null,"url":null,"abstract":"A 3D virtual view synthesis system is to generate a virtual view at an arbitrary viewpoint from the texture and depth information of multiple reference views. There are several technical challenges in producing a high-quality synthesized view such as warping, blending, ghost artifact reduction and hole filling. Four tools (techniques) have been proposed in this paper to solve these problems. They are unreliable region extraction, fast backward warping, adaptive blending, and hole filling by using the constructed background models. Other techniques such as disocclusion detection, bicubic interpolation, and background model construction are also employed in the proposed algorithm. All above techniques have been designed and tested on the MPEG test sequences. Experimental results show that high-quality virtual views are generated with fewer artifacts.","PeriodicalId":378956,"journal":{"name":"2016 3DTV-Conference: The True Vision - Capture, Transmission and Display of 3D Video (3DTV-CON)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2016-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"Virtual view synthesis quality refinement\",\"authors\":\"Tzu-Chin Lee, Chun-Liang Chien, H. Hang\",\"doi\":\"10.1109/3DTV.2016.7548958\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"A 3D virtual view synthesis system is to generate a virtual view at an arbitrary viewpoint from the texture and depth information of multiple reference views. There are several technical challenges in producing a high-quality synthesized view such as warping, blending, ghost artifact reduction and hole filling. Four tools (techniques) have been proposed in this paper to solve these problems. They are unreliable region extraction, fast backward warping, adaptive blending, and hole filling by using the constructed background models. Other techniques such as disocclusion detection, bicubic interpolation, and background model construction are also employed in the proposed algorithm. All above techniques have been designed and tested on the MPEG test sequences. Experimental results show that high-quality virtual views are generated with fewer artifacts.\",\"PeriodicalId\":378956,\"journal\":{\"name\":\"2016 3DTV-Conference: The True Vision - Capture, Transmission and Display of 3D Video (3DTV-CON)\",\"volume\":\"4 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2016-07-04\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2016 3DTV-Conference: The True Vision - Capture, Transmission and Display of 3D Video (3DTV-CON)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/3DTV.2016.7548958\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2016 3DTV-Conference: The True Vision - Capture, Transmission and Display of 3D Video (3DTV-CON)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/3DTV.2016.7548958","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
A 3D virtual view synthesis system is to generate a virtual view at an arbitrary viewpoint from the texture and depth information of multiple reference views. There are several technical challenges in producing a high-quality synthesized view such as warping, blending, ghost artifact reduction and hole filling. Four tools (techniques) have been proposed in this paper to solve these problems. They are unreliable region extraction, fast backward warping, adaptive blending, and hole filling by using the constructed background models. Other techniques such as disocclusion detection, bicubic interpolation, and background model construction are also employed in the proposed algorithm. All above techniques have been designed and tested on the MPEG test sequences. Experimental results show that high-quality virtual views are generated with fewer artifacts.