Learning CRFs Using Graph Cuts

Szummer, Martin; Kohli, Pushmeet; Hoiem, Derek

doi:10.1007/978-3-540-88688-4_43

Cited by 187 publications

(225 citation statements)

References 14 publications

Supporting

Mentioning

225

Contrasting

Order By: Relevance

“…Before discussing its benefits to organizations, it is necessary to review the contents of interactive learning and how industry might make best use of them. Learning activities for interactive learning is essential for training and learning since the efficiency can be improved (Chang, 2003;Szummer et al, 2008). Learning activities are co-ordinated actions that exercise basic intellectual skills, thought processes, and analysis techniques.…”

Section: Interactive Learningmentioning

confidence: 99%

Review and discussion: E-learning for academia and industry

Chang

2016

International Journal of Information Management

190

118

View full text Add to dashboard Cite

This paper presents a high level review and discussion about e-learning and proposes the use of interactive learning as a recommended method for staff training in industry and academia. Interactive learning is focused on the integrated e-learning and face-to-face learning to ensure that the process of learning can stimulate learners' interests, report their progress and have tutors to provide their feedback and guide learners to the expected targets. Learning activities and varieties have been illustrated with discussion about how industry and academia can use interactive learning. Five successful examples of interactive learning to demonstrate the effectiveness of interactive learning. Positive impacts have been reported in RBS, SMEs using SAP, University of Cambridge, University of Greenwich and Leeds Beckett University to support the positive outcomes for learners and trainers. Future directions have been discussed, particularly the use of emerging services can enhance the learning experience and satisfaction for learners and trainers.

show abstract

Section: Interactive Learningmentioning

confidence: 99%

Review and discussion: E-learning for academia and industry

Chang

2016

International Journal of Information Management

190

118

View full text Add to dashboard Cite

show abstract

“…(6) is an NP-hard problem. Referring to the max-margin CRF learning approach, 44 we look for the optimal weights υ and codebook C that assign the training labels Y ðkÞ , a probability that is greater than or equal to any other labeling Y of instance k E Q -T A R G E T ; t e m p : i n t r a l i n k -; e 0 1 1 ; 3 2 6 ; 4 0 2 P½Y ðkÞ jL ðkÞ ; υ ≥ P½YjL ðkÞ ; υ ∀Y \ Y ðkÞ ∀k:…”

Section: Joint Crf and Codebook Learningmentioning

confidence: 99%

Improved Hough transform by modeling context with conditional random fields for partially occluded pedestrian detection

Jiang

Xiong

2018

Opt. Eng.

View full text Add to dashboard Cite

Abstract. Traditional Hough transform-based methods detect objects by casting votes to object centroids from object patches. It is difficult to disambiguate object patches from the background by a classifier without contextual information, as an image patch only carries partial information about the object. To leverage the contextual information among image patches, we capture the contextual relationships on image patches through a conditional random field (CRF) with latent variables denoted by locality-constrained linear coding (LLC). The strength of the pairwise energy in the CRF is measured using a Gaussian kernel. In the training stage, we modulate the visual codebook by learning the CRF model iteratively. In the test stage, the binary labels of image patches are jointly estimated by the CRF model. Image patches labeled as the object category cast weighted votes for object centroids in an image according to the LLC coefficients. Experimental results on the INRIA pedestrian, TUD Brussels, and Caltech pedestrian datasets demonstrate the effectiveness of the proposed method compared with other Hough transform-based methods. © The Authors. Published by SPIE under a Creative Commons Attribution 3.0 Unported License. Distribution or reproduction of this work in whole or in part requires full attribution of the original publication, including its DOI.

show abstract

“…Li and Huttenlocher [19] proposed an SSVM-based algorithm in the context of stereo. Szummer et al [31] used graph-cuts within an SSVM learning algorithm in the context of segmentation. We note that in both cases, the energies of the models were linear in the parameters.…”

Section: Related Workmentioning

confidence: 99%

Learning the right model: Efficient max-margin learning in Laplacian CRFs

Batra

Saxena

2012

2012 IEEE Conference on Computer Vision and Pattern Recognition

View full text Add to dashboard Cite

An important modeling decision made while designing Conditional Random Fields (CRFs) is the choice of the potential functions over the cliques of variables. Laplacian potentials are useful because they are robust potentials and match image statistics better than Gaussians. Moreover, energies with Laplacian terms remain convex, which simplifies inference. This makes Laplacian potentials an ideal modeling choice for some applications.In this paper, we study max-margin parameter learning in CRFs with Laplacian potentials (LCRFs). We first show that structured hinge-loss [35] is non-convex for LCRFs and thus techniques used by previous works are not applicable. We then present the first approximate max-margin algorithm for LCRFs. Finally, we make our learning algorithm scalable in the number of training images by using dual-decomposition techniques. Our experiments on singleimage depth estimation show that even with simple features, our approach achieves comparable to state-of-art results.

show abstract

Learning CRFs Using Graph Cuts

Cited by 187 publications

References 14 publications

Review and discussion: E-learning for academia and industry

Review and discussion: E-learning for academia and industry

Improved Hough transform by modeling context with conditional random fields for partially occluded pedestrian detection

Learning the right model: Efficient max-margin learning in Laplacian CRFs

Contact Info

Product

Resources

About