2D-3D Lifting for Shape Reconstruction
Nan, Liangliang; Sharf, Andrei; Chen, Baoquan
刊名COMPUTER GRAPHICS FORUM
2014
英文摘要We present an algorithm for shape reconstruction from incomplete 3D scans by fusing together two acquisition modes: 2D photographs and 3D scans. The two modes exhibit complementary characteristics: scans have depth information, but are often sparse and incomplete; photographs, on the other hand, are dense and have high resolution, but lack important depth information. In this work we fuse the two modes, taking advantage of their complementary information, to enhance 3D shape reconstruction from an incomplete scan with a 2D photograph. We compute geometrical and topological shape properties in 2Dphotographs and use them to reconstruct a shape from an incomplete 3D scan in a principled manner. Our key observation is that shape properties such as boundaries, smooth patches and local connectivity, can be inferred with high confidence from 2D photographs. Thus, we register the 3D scan with the 2Dphotograph and use scanned points as 3D depth cues for lifting 2D shape structures into 3D. Our contribution is an algorithm which significantly regularizes and enhances the problem of 3D reconstruction from partial scans by lifting 2D shape structures into 3D. We evaluate our algorithm on various shapes which are loosely scanned and photographed from different views, and compare them with state-of-the-art reconstruction methods.
收录类别SCI
原文出处http://onlinelibrary.wiley.com/doi/10.1111/cgf.12493/abstract
语种英语
内容类型期刊论文
源URL[http://ir.siat.ac.cn:8080/handle/172644/5981]  
专题深圳先进技术研究院_数字所
作者单位COMPUTER GRAPHICS FORUM
推荐引用方式
GB/T 7714
Nan, Liangliang,Sharf, Andrei,Chen, Baoquan. 2D-3D Lifting for Shape Reconstruction[J]. COMPUTER GRAPHICS FORUM,2014.
APA Nan, Liangliang,Sharf, Andrei,&Chen, Baoquan.(2014).2D-3D Lifting for Shape Reconstruction.COMPUTER GRAPHICS FORUM.
MLA Nan, Liangliang,et al."2D-3D Lifting for Shape Reconstruction".COMPUTER GRAPHICS FORUM (2014).
个性服务
查看访问统计
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。


©版权所有 ©2017 CSpace - Powered by CSpace