G2MF-WA: Geometric multi-model fitting with weakly annotated data
Version 2 2024-06-13, 07:34Version 2 2024-06-13, 07:34
Version 1 2020-04-22, 09:26Version 1 2020-04-22, 09:26
journal contribution
posted on 2024-06-13, 07:34 authored by C Zhang, X Lu, K Hotta, X Yang© 2020, The Author(s). In this paper we address the problem of geometric multi-model fitting using a few weakly annotated data points, which has been little studied so far. In weak annotating (WA), most manual annotations are supposed to be correct yet inevitably mixed with incorrect ones. SuchWA data can naturally arise through interaction in various tasks. For example, in the case of homography estimation, one can easily annotate points on the same plane or object with a single label by observing the image. Motivated by this, we propose a novel method to make full use of WA data to boost multi-model fitting performance. Specifically, a graph for model proposal sampling is first constructed using the WA data, given the prior that WA data annotated with the same weak label has a high probability of belonging to the same model. By incorporating this prior knowledge into the calculation of edge probabilities, vertices (i.e., data points) lying on or near the latent model are likely to be associated and further form a subset or cluster for effective proposal generation. Having generated proposals, α-expansion is used for labeling, and our method in return updates the proposals. This procedure works in an iterative way. Extensive experiments validate our method and show that it produces noticeably better results than state-of-the-art techniques in most cases.
History
Journal
Computational Visual MediaVolume
6Pagination
135-145Location
Berlin, GermanyPublisher DOI
Link to full text
ISSN
2096-0433eISSN
2096-0662Language
engPublication classification
C1 Refereed article in a scholarly journalCopyright notice
2020, The Author(s)Issue
2Publisher
SpringerUsage metrics
Categories
Keywords
Licence
Exports
RefWorksRefWorks
BibTeXBibTeX
Ref. managerRef. manager
EndnoteEndnote
DataCiteDataCite
NLMNLM
DCDC