Multiview reconstruction of space curves
(2003) IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION, 2003 In Proceedings of the IEEE International Conference on Computer Vision 2. p.10171024 Abstract
 Is the real problem in resolving correspondence using current stereo algorithms the lack of the "right" matching criterion? In studying the related task of reconstructing threedimensional space curves from their projections in multiple views, we suggest that the problem is more basic: matching and reconstruction are coupled, and so reconstruction algorithms should exploit ths rather than assuming that matching can be successfully performed before reconstruction. To realize this coupling, a generative model of curves is introduced which has two key components: (i) a prior distribution of general space curves and (ii) an image formation model whch discribes how 3D curves are projected onto the image plane. A novel aspect of the image... (More)
 Is the real problem in resolving correspondence using current stereo algorithms the lack of the "right" matching criterion? In studying the related task of reconstructing threedimensional space curves from their projections in multiple views, we suggest that the problem is more basic: matching and reconstruction are coupled, and so reconstruction algorithms should exploit ths rather than assuming that matching can be successfully performed before reconstruction. To realize this coupling, a generative model of curves is introduced which has two key components: (i) a prior distribution of general space curves and (ii) an image formation model whch discribes how 3D curves are projected onto the image plane. A novel aspect of the image formation model is that it uses an exact description of the gradient field of a piecewise constant image. Based on this forward model, a fully automatic algorithm for solving the inverse problem is developed for an arbitrary number of views. The resulting algorithm is robust to partial occlusion, deficiencies in image curve extraction and it does not rely on photometric information. The relative motion of the cameras is assumed to be given. Several experiments are carried out on various realistic scenarios. In particular, we focus on scenes where traditional correlationbased methods would fail. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/record/611933
 author
 Kahl, Fredrik ^{LU} and August, Jonas
 organization
 publishing date
 2003
 type
 Chapter in Book/Report/Conference proceeding
 publication status
 published
 subject
 keywords
 Stereo algorithms, Piecewise constant image, Curve extraction
 in
 Proceedings of the IEEE International Conference on Computer Vision
 volume
 2
 pages
 1017  1024
 publisher
 IEEEInstitute of Electrical and Electronics Engineers Inc.
 conference name
 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION, 2003
 external identifiers

 WOS:000186833000135
 Other:CODEN: PICVES
 Scopus:0344120679
 DOI
 10.1109/ICCV.2003.1238461
 language
 English
 LU publication?
 yes
 id
 f8c385245f9c4d839fd71ce679bffe87 (old id 611933)
 date added to LUP
 20071130 11:50:24
 date last changed
 20170108 05:37:23
@inproceedings{f8c385245f9c4d839fd71ce679bffe87, abstract = {Is the real problem in resolving correspondence using current stereo algorithms the lack of the "right" matching criterion? In studying the related task of reconstructing threedimensional space curves from their projections in multiple views, we suggest that the problem is more basic: matching and reconstruction are coupled, and so reconstruction algorithms should exploit ths rather than assuming that matching can be successfully performed before reconstruction. To realize this coupling, a generative model of curves is introduced which has two key components: (i) a prior distribution of general space curves and (ii) an image formation model whch discribes how 3D curves are projected onto the image plane. A novel aspect of the image formation model is that it uses an exact description of the gradient field of a piecewise constant image. Based on this forward model, a fully automatic algorithm for solving the inverse problem is developed for an arbitrary number of views. The resulting algorithm is robust to partial occlusion, deficiencies in image curve extraction and it does not rely on photometric information. The relative motion of the cameras is assumed to be given. Several experiments are carried out on various realistic scenarios. In particular, we focus on scenes where traditional correlationbased methods would fail.}, author = {Kahl, Fredrik and August, Jonas}, booktitle = {Proceedings of the IEEE International Conference on Computer Vision}, keyword = {Stereo algorithms,Piecewise constant image,Curve extraction}, language = {eng}, pages = {10171024}, publisher = {IEEEInstitute of Electrical and Electronics Engineers Inc.}, title = {Multiview reconstruction of space curves}, url = {http://dx.doi.org/10.1109/ICCV.2003.1238461}, volume = {2}, year = {2003}, }