CASIA OpenIR  > 模式识别国家重点实验室  > 多媒体计算与图形学
PM-PM: PatchMatch With Potts Model for Object Segmentation and Stereo Matching
Xu, Shibiao1,2; Zhang, Feihu2; He, Xiaofei3; Shen, Xukun1; Zhang, Xiaopeng2
Source PublicationIEEE TRANSACTIONS ON IMAGE PROCESSING
2015-03-25
Volume24Issue:7Pages:2182-2196
SubtypeArticle
AbstractThis paper presents a unified variational formulation for joint object segmentation and stereo matching, which takes both accuracy and efficiency into account. In our approach, depth-map consists of compact objects, each object is represented through three different aspects: 1) the perimeter in image space; 2) the slanted object depth plane; and 3) the planar bias, which is to add an additional level of detail on top of each object plane in order to model depth variations within an object. Compared with traditional high quality solving methods in low level, we use a convex formulation of the multilabel Potts Model with PatchMatch stereo techniques to generate depth-map at each image in object level and show that accurate multiple view reconstruction can be achieved with our formulation by means of induced homography without discretization or staircasing artifacts. Our model is formulated as an energy minimization that is optimized via a fast primal-dual algorithm, which can handle several hundred object depth segments efficiently. Performance evaluations in the Middlebury benchmark data sets show that our method outperforms the traditional integer-valued disparity strategy as well as the original PatchMatch algorithm and its variants in subpixel accurate disparity estimation. The proposed algorithm is also evaluated and shown to produce consistently good results for various real-world data sets (KITTI benchmark data sets and multiview benchmark data sets).
KeywordObject Segmentation Stereo Matching Potts Model Patchmatch Multiple View Reconstruction
WOS HeadingsScience & Technology ; Technology
WOS KeywordCORRESPONDENCE FIELD ESTIMATION ; BELIEF PROPAGATION ; IMAGE SEGMENTATION ; GRAPH CUTS ; COST AGGREGATION ; OPTIMIZATION ; MINIMIZATION ; ALGORITHMS ; SEARCH ; COLOR
Indexed BySCI
Language英语
WOS Research AreaComputer Science ; Engineering
WOS SubjectComputer Science, Artificial Intelligence ; Engineering, Electrical & Electronic
WOS IDWOS:000352732800006
Citation statistics
Cited Times:34[WOS]   [WOS Record]     [Related Records in WOS]
Document Type期刊论文
Identifierhttp://ir.ia.ac.cn/handle/173211/8106
Collection模式识别国家重点实验室_多媒体计算与图形学
Affiliation1.Beihang Univ, State Key Lab Virtual Real Technol & Syst, Beijing 100191, Peoples R China
2.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
3.Zhejiang Univ, Coll Comp Sci, Hangzhou 310027, Zhejiang, Peoples R China
Recommended Citation
GB/T 7714
Xu, Shibiao,Zhang, Feihu,He, Xiaofei,et al. PM-PM: PatchMatch With Potts Model for Object Segmentation and Stereo Matching[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING,2015,24(7):2182-2196.
APA Xu, Shibiao,Zhang, Feihu,He, Xiaofei,Shen, Xukun,&Zhang, Xiaopeng.(2015).PM-PM: PatchMatch With Potts Model for Object Segmentation and Stereo Matching.IEEE TRANSACTIONS ON IMAGE PROCESSING,24(7),2182-2196.
MLA Xu, Shibiao,et al."PM-PM: PatchMatch With Potts Model for Object Segmentation and Stereo Matching".IEEE TRANSACTIONS ON IMAGE PROCESSING 24.7(2015):2182-2196.
Files in This Item: Download All
File Name/Size DocType Version Access License
PM-PM_PatchMatch wit(5719KB)期刊论文出版稿开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Xu, Shibiao]'s Articles
[Zhang, Feihu]'s Articles
[He, Xiaofei]'s Articles
Baidu academic
Similar articles in Baidu academic
[Xu, Shibiao]'s Articles
[Zhang, Feihu]'s Articles
[He, Xiaofei]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Xu, Shibiao]'s Articles
[Zhang, Feihu]'s Articles
[He, Xiaofei]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: PM-PM_PatchMatch with Potts model for object segmentation and stereo matching.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.