{"title":"基于L0梯度最小化的单幅图像深度恢复","authors":"Fengyun Cao, Fei Xie","doi":"10.1109/ICBK.2018.00052","DOIUrl":null,"url":null,"abstract":"Aiming at the challenging problem of single image depth recovery, a new local defocus blur estimation algorithm is presented based on L0 gradient minimization. There is a common problem in the existing methods, that is, quantization error at weak edges, noise or soft shadows may cause inaccurate blur estimates at some edge locations. In the proposed methods, L0 smoothing technology is employed to screen the effective edge which is advantageous to estimate defocus information and denoising. The guided image filter is applied to propagate the blur amount at edge locations to the entire image, a refined defocus map can be obtained. Then the recovery of the relative depth order on the image is achieved from the blur map. At last, T-junction is adopted to eliminate the ambiguity in the depth map over the focal plane. Experimental results demonstrate that compared with the previous approaches, the algorithm can effectively produces high quality depth maps.","PeriodicalId":144958,"journal":{"name":"2018 IEEE International Conference on Big Knowledge (ICBK)","volume":"97 ","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Depth Recovery from a Single Image Based on L0 Gradient Minimization\",\"authors\":\"Fengyun Cao, Fei Xie\",\"doi\":\"10.1109/ICBK.2018.00052\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Aiming at the challenging problem of single image depth recovery, a new local defocus blur estimation algorithm is presented based on L0 gradient minimization. There is a common problem in the existing methods, that is, quantization error at weak edges, noise or soft shadows may cause inaccurate blur estimates at some edge locations. In the proposed methods, L0 smoothing technology is employed to screen the effective edge which is advantageous to estimate defocus information and denoising. The guided image filter is applied to propagate the blur amount at edge locations to the entire image, a refined defocus map can be obtained. Then the recovery of the relative depth order on the image is achieved from the blur map. At last, T-junction is adopted to eliminate the ambiguity in the depth map over the focal plane. Experimental results demonstrate that compared with the previous approaches, the algorithm can effectively produces high quality depth maps.\",\"PeriodicalId\":144958,\"journal\":{\"name\":\"2018 IEEE International Conference on Big Knowledge (ICBK)\",\"volume\":\"97 \",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2018-11-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2018 IEEE International Conference on Big Knowledge (ICBK)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICBK.2018.00052\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2018 IEEE International Conference on Big Knowledge (ICBK)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICBK.2018.00052","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Depth Recovery from a Single Image Based on L0 Gradient Minimization
Aiming at the challenging problem of single image depth recovery, a new local defocus blur estimation algorithm is presented based on L0 gradient minimization. There is a common problem in the existing methods, that is, quantization error at weak edges, noise or soft shadows may cause inaccurate blur estimates at some edge locations. In the proposed methods, L0 smoothing technology is employed to screen the effective edge which is advantageous to estimate defocus information and denoising. The guided image filter is applied to propagate the blur amount at edge locations to the entire image, a refined defocus map can be obtained. Then the recovery of the relative depth order on the image is achieved from the blur map. At last, T-junction is adopted to eliminate the ambiguity in the depth map over the focal plane. Experimental results demonstrate that compared with the previous approaches, the algorithm can effectively produces high quality depth maps.