Nonparametric image parsing using adaptive neighbor sets

David Eigen, Rob Fergus

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper proposes a non-parametric approach to scene parsing inspired by the work of Tighe and Lazebnik [22]. In their approach, a simple kNN scheme with multiple descriptor types is used to classify super-pixels. We add two novel mechanisms: (i) a principled and efficient method for learning per-descriptor weights that minimizes classification error, and (ii) a context-driven adaptation of the training set used for each query, which conditions on common classes (which are relatively easy to classify) to improve performance on rare ones. The first technique helps to remove extraneous descriptors that result from the imperfect distance metrics/representations of each super-pixel. The second contribution re-balances the class frequencies, away from the highly-skewed distribution found in real-world scenes. Both methods give a significant performance boost over [22] and the overall system achieves state-of-the-art performance on the SIFT-Flow dataset.

Original languageEnglish (US)
Title of host publication2012 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2012
Pages2799-2806
Number of pages8
DOIs
StatePublished - 2012
Event2012 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2012 - Providence, RI, United States
Duration: Jun 16 2012Jun 21 2012

Other

Other2012 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2012
CountryUnited States
CityProvidence, RI
Period6/16/126/21/12

Fingerprint

Pixels

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition

Cite this

Eigen, D., & Fergus, R. (2012). Nonparametric image parsing using adaptive neighbor sets. In 2012 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2012 (pp. 2799-2806). [6248004] https://doi.org/10.1109/CVPR.2012.6248004

Nonparametric image parsing using adaptive neighbor sets. / Eigen, David; Fergus, Rob.

2012 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2012. 2012. p. 2799-2806 6248004.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Eigen, D & Fergus, R 2012, Nonparametric image parsing using adaptive neighbor sets. in 2012 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2012., 6248004, pp. 2799-2806, 2012 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2012, Providence, RI, United States, 6/16/12. https://doi.org/10.1109/CVPR.2012.6248004
Eigen D, Fergus R. Nonparametric image parsing using adaptive neighbor sets. In 2012 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2012. 2012. p. 2799-2806. 6248004 https://doi.org/10.1109/CVPR.2012.6248004
Eigen, David ; Fergus, Rob. / Nonparametric image parsing using adaptive neighbor sets. 2012 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2012. 2012. pp. 2799-2806
@inproceedings{efaff41f83a247a981d1ffca4e7c650b,
title = "Nonparametric image parsing using adaptive neighbor sets",
abstract = "This paper proposes a non-parametric approach to scene parsing inspired by the work of Tighe and Lazebnik [22]. In their approach, a simple kNN scheme with multiple descriptor types is used to classify super-pixels. We add two novel mechanisms: (i) a principled and efficient method for learning per-descriptor weights that minimizes classification error, and (ii) a context-driven adaptation of the training set used for each query, which conditions on common classes (which are relatively easy to classify) to improve performance on rare ones. The first technique helps to remove extraneous descriptors that result from the imperfect distance metrics/representations of each super-pixel. The second contribution re-balances the class frequencies, away from the highly-skewed distribution found in real-world scenes. Both methods give a significant performance boost over [22] and the overall system achieves state-of-the-art performance on the SIFT-Flow dataset.",
author = "David Eigen and Rob Fergus",
year = "2012",
doi = "10.1109/CVPR.2012.6248004",
language = "English (US)",
isbn = "9781467312264",
pages = "2799--2806",
booktitle = "2012 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2012",

}

TY - GEN

T1 - Nonparametric image parsing using adaptive neighbor sets

AU - Eigen, David

AU - Fergus, Rob

PY - 2012

Y1 - 2012

N2 - This paper proposes a non-parametric approach to scene parsing inspired by the work of Tighe and Lazebnik [22]. In their approach, a simple kNN scheme with multiple descriptor types is used to classify super-pixels. We add two novel mechanisms: (i) a principled and efficient method for learning per-descriptor weights that minimizes classification error, and (ii) a context-driven adaptation of the training set used for each query, which conditions on common classes (which are relatively easy to classify) to improve performance on rare ones. The first technique helps to remove extraneous descriptors that result from the imperfect distance metrics/representations of each super-pixel. The second contribution re-balances the class frequencies, away from the highly-skewed distribution found in real-world scenes. Both methods give a significant performance boost over [22] and the overall system achieves state-of-the-art performance on the SIFT-Flow dataset.

AB - This paper proposes a non-parametric approach to scene parsing inspired by the work of Tighe and Lazebnik [22]. In their approach, a simple kNN scheme with multiple descriptor types is used to classify super-pixels. We add two novel mechanisms: (i) a principled and efficient method for learning per-descriptor weights that minimizes classification error, and (ii) a context-driven adaptation of the training set used for each query, which conditions on common classes (which are relatively easy to classify) to improve performance on rare ones. The first technique helps to remove extraneous descriptors that result from the imperfect distance metrics/representations of each super-pixel. The second contribution re-balances the class frequencies, away from the highly-skewed distribution found in real-world scenes. Both methods give a significant performance boost over [22] and the overall system achieves state-of-the-art performance on the SIFT-Flow dataset.

UR - http://www.scopus.com/inward/record.url?scp=84866674722&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84866674722&partnerID=8YFLogxK

U2 - 10.1109/CVPR.2012.6248004

DO - 10.1109/CVPR.2012.6248004

M3 - Conference contribution

SN - 9781467312264

SP - 2799

EP - 2806

BT - 2012 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2012

ER -