Max-margin Boltzmann machines for object segmentation

Jimei Yang; Simon Sáfár; Ming Hsuan Yang

doi:10.1109/CVPR.2014.48

Max-margin Boltzmann machines for object segmentation

Jimei Yang, Simon Sáfár, Ming Hsuan Yang

Department of Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

21 Citations (Scopus)

Abstract

We present Max-Margin Boltzmann Machines (MMBMs) for object segmentation. MMBMs are essentially a class of Conditional Boltzmann Machines that model the joint distribution of hidden variables and output labels conditioned on input observations. In addition to image-to-label connections, we build direct image-to-hidden connections to facilitate global shape prediction, and thus derive a simple Iterated Conditional Modes algorithm for efficient maximum a posteriori inference. We formulate a max-margin objective function for discriminative training, and analyze the effects of different margin functions on learning. We evaluate MMBMs using three datasets against state-of-the-art methods to demonstrate the strength of the proposed algorithms.

Original language	English
Title of host publication	Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
Publisher	IEEE Computer Society
Pages	320-327
Number of pages	8
ISBN (Electronic)	9781479951178, 9781479951178
DOIs	https://doi.org/10.1109/CVPR.2014.48
Publication status	Published - 2014 Sep 24
Event	27th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2014 - Columbus, United States Duration: 2014 Jun 23 → 2014 Jun 28

Publication series

Name	Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
ISSN (Print)	1063-6919

Other

Other	27th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2014
Country	United States
City	Columbus
Period	14/6/23 → 14/6/28

All Science Journal Classification (ASJC) codes

Software
Computer Vision and Pattern Recognition

Access to Document

10.1109/CVPR.2014.48

Fingerprint Dive into the research topics of 'Max-margin Boltzmann machines for object segmentation'. Together they form a unique fingerprint.

Cite this

Yang, J., Sáfár, S., & Yang, M. H. (2014). Max-margin Boltzmann machines for object segmentation. In Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (pp. 320-327). [6909442] (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition). IEEE Computer Society. https://doi.org/10.1109/CVPR.2014.48

@inproceedings{d268240b2d9340ffbc3ebe723e93df61,

title = "Max-margin Boltzmann machines for object segmentation",

abstract = "We present Max-Margin Boltzmann Machines (MMBMs) for object segmentation. MMBMs are essentially a class of Conditional Boltzmann Machines that model the joint distribution of hidden variables and output labels conditioned on input observations. In addition to image-to-label connections, we build direct image-to-hidden connections to facilitate global shape prediction, and thus derive a simple Iterated Conditional Modes algorithm for efficient maximum a posteriori inference. We formulate a max-margin objective function for discriminative training, and analyze the effects of different margin functions on learning. We evaluate MMBMs using three datasets against state-of-the-art methods to demonstrate the strength of the proposed algorithms.",

author = "Jimei Yang and Simon S{\'a}f{\'a}r and Yang, {Ming Hsuan}",

note = "Publisher Copyright: {\textcopyright} 2014 IEEE. Copyright: Copyright 2015 Elsevier B.V., All rights reserved.; 27th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2014 ; Conference date: 23-06-2014 Through 28-06-2014",

year = "2014",

month = sep,

day = "24",

doi = "10.1109/CVPR.2014.48",

language = "English",

series = "Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition",

publisher = "IEEE Computer Society",

pages = "320--327",

booktitle = "Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition",

address = "United States",

}

Yang, J, Sáfár, S & Yang, MH 2014, Max-margin Boltzmann machines for object segmentation. in Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition., 6909442, Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, IEEE Computer Society, pp. 320-327, 27th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2014, Columbus, United States, 14/6/23. https://doi.org/10.1109/CVPR.2014.48

Max-margin Boltzmann machines for object segmentation. / Yang, Jimei; Sáfár, Simon; Yang, Ming Hsuan.

Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. IEEE Computer Society, 2014. p. 320-327 6909442 (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Max-margin Boltzmann machines for object segmentation

AU - Yang, Jimei

AU - Sáfár, Simon

AU - Yang, Ming Hsuan

PY - 2014/9/24

Y1 - 2014/9/24

N2 - We present Max-Margin Boltzmann Machines (MMBMs) for object segmentation. MMBMs are essentially a class of Conditional Boltzmann Machines that model the joint distribution of hidden variables and output labels conditioned on input observations. In addition to image-to-label connections, we build direct image-to-hidden connections to facilitate global shape prediction, and thus derive a simple Iterated Conditional Modes algorithm for efficient maximum a posteriori inference. We formulate a max-margin objective function for discriminative training, and analyze the effects of different margin functions on learning. We evaluate MMBMs using three datasets against state-of-the-art methods to demonstrate the strength of the proposed algorithms.

AB - We present Max-Margin Boltzmann Machines (MMBMs) for object segmentation. MMBMs are essentially a class of Conditional Boltzmann Machines that model the joint distribution of hidden variables and output labels conditioned on input observations. In addition to image-to-label connections, we build direct image-to-hidden connections to facilitate global shape prediction, and thus derive a simple Iterated Conditional Modes algorithm for efficient maximum a posteriori inference. We formulate a max-margin objective function for discriminative training, and analyze the effects of different margin functions on learning. We evaluate MMBMs using three datasets against state-of-the-art methods to demonstrate the strength of the proposed algorithms.

UR - http://www.scopus.com/inward/record.url?scp=84911438661&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84911438661&partnerID=8YFLogxK

U2 - 10.1109/CVPR.2014.48

DO - 10.1109/CVPR.2014.48

M3 - Conference contribution

AN - SCOPUS:84911438661

T3 - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition

SP - 320

EP - 327

BT - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition

PB - IEEE Computer Society

T2 - 27th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2014

Y2 - 23 June 2014 through 28 June 2014

ER -