Please use this identifier to cite or link to this item:
Scopus Web of Science® Altmetric
Type: Conference paper
Title: Superpixel-based two-view deterministic fitting for multiple-structure data
Author: Xiao, G.
Wang, H.
Yan, Y.
Suter, D.
Citation: Lecture Notes in Artificial Intelligence, 2016 / Leibe, B., Matas, J., Sebe, N., Welling, M. (ed./s), vol.9910, pp.517-533
Publisher: Springer International Publishing AG
Issue Date: 2016
Series/Report no.: Lecture Notes in Computer Science
ISBN: 9783319464657
ISSN: 0302-9743
Conference Name: 14th European Conference on Computer Vision (ECCV) (8 Oct 2016 - 16 Oct 2016 : Amsterdam, Netherlands)
Editor: Leibe, B.
Matas, J.
Sebe, N.
Welling, M.
Statement of
Guobao Xiao, Hanzi Wang, B, Yan Yan, and David Suter
Abstract: This paper proposes a two-view deterministic geometric model fitting method, termed Superpixel-based Deterministic Fitting (SDF), for multiple-structure data. SDF starts from superpixel segmentation, which effectively captures prior information of feature appearances. The feature appearances are beneficial to reduce the computational complexity for deterministic fitting methods. SDF also includes two original elements, i.e., a deterministic sampling algorithm and a novel model selection algorithm. The two algorithms are tightly coupled to boost the performance of SDF in both speed and accuracy. The key characteristic of SDF is that it can efficiently and deterministically estimate the parameters of model instances in multi-structure data. Experimental results demonstrate that the proposed SDF shows superiority over several state-of-the-art fitting methods for real images with singlestructure and multiple-structure data.
Keywords: Deterministic algorithm; Superpixel; Model fitting; Feature appearances
Rights: © Springer International Publishing AG 2016
DOI: 10.1007/978-3-319-46466-4_31
Grant ID:
Published version:
Appears in Collections:Aurora harvest 3
Computer Science publications

Files in This Item:
File Description SizeFormat 
  Restricted Access
Restricted Access4.6 MBAdobe PDFView/Open

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