A 3D feature model for image matching

Zachary Sun, Nadya Bliss, Karl Ni

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

2 Citations (Scopus)

Abstract

The proposed algorithm identifies whether or not a test photo belongs to a set of co-located training images based on its spatial proximity to the training set. We leverage concepts from Lowe's SIFT and Snavely's Photo Tourism algorithms, and match an image by its 2D features to the 3D features representing the training set. To reduce complexity and increase efficiency, the proposed algorithm implements a compact representation of the image set by merging collections similar features. Test images are then matched with the derived structure. Finally, a decision statistic is determined based on the percentage of features that match. Receiver operating characteristics, computational analysis, and distributions are included in the performance analysis.

Original languageEnglish (US)
Title of host publicationICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Pages2194-2197
Number of pages4
DOIs
StatePublished - 2010
Externally publishedYes
Event2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010 - Dallas, TX, United States
Duration: Mar 14 2010Mar 19 2010

Other

Other2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010
CountryUnited States
CityDallas, TX
Period3/14/103/19/10

Fingerprint

Image matching
Merging
Statistics

Keywords

  • 3D matching
  • Approximate nearest neighbor
  • SIFT
  • Structure from motion

ASJC Scopus subject areas

  • Signal Processing
  • Software
  • Electrical and Electronic Engineering

Cite this

Sun, Z., Bliss, N., & Ni, K. (2010). A 3D feature model for image matching. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings (pp. 2194-2197). [5495705] https://doi.org/10.1109/ICASSP.2010.5495705

A 3D feature model for image matching. / Sun, Zachary; Bliss, Nadya; Ni, Karl.

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2010. p. 2194-2197 5495705.

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

Sun, Z, Bliss, N & Ni, K 2010, A 3D feature model for image matching. in ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings., 5495705, pp. 2194-2197, 2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010, Dallas, TX, United States, 3/14/10. https://doi.org/10.1109/ICASSP.2010.5495705
Sun Z, Bliss N, Ni K. A 3D feature model for image matching. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2010. p. 2194-2197. 5495705 https://doi.org/10.1109/ICASSP.2010.5495705
Sun, Zachary ; Bliss, Nadya ; Ni, Karl. / A 3D feature model for image matching. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2010. pp. 2194-2197
@inproceedings{b35e24bfa767409aa1ae61251e656db3,
title = "A 3D feature model for image matching",
abstract = "The proposed algorithm identifies whether or not a test photo belongs to a set of co-located training images based on its spatial proximity to the training set. We leverage concepts from Lowe's SIFT and Snavely's Photo Tourism algorithms, and match an image by its 2D features to the 3D features representing the training set. To reduce complexity and increase efficiency, the proposed algorithm implements a compact representation of the image set by merging collections similar features. Test images are then matched with the derived structure. Finally, a decision statistic is determined based on the percentage of features that match. Receiver operating characteristics, computational analysis, and distributions are included in the performance analysis.",
keywords = "3D matching, Approximate nearest neighbor, SIFT, Structure from motion",
author = "Zachary Sun and Nadya Bliss and Karl Ni",
year = "2010",
doi = "10.1109/ICASSP.2010.5495705",
language = "English (US)",
isbn = "9781424442966",
pages = "2194--2197",
booktitle = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

}

TY - GEN

T1 - A 3D feature model for image matching

AU - Sun, Zachary

AU - Bliss, Nadya

AU - Ni, Karl

PY - 2010

Y1 - 2010

N2 - The proposed algorithm identifies whether or not a test photo belongs to a set of co-located training images based on its spatial proximity to the training set. We leverage concepts from Lowe's SIFT and Snavely's Photo Tourism algorithms, and match an image by its 2D features to the 3D features representing the training set. To reduce complexity and increase efficiency, the proposed algorithm implements a compact representation of the image set by merging collections similar features. Test images are then matched with the derived structure. Finally, a decision statistic is determined based on the percentage of features that match. Receiver operating characteristics, computational analysis, and distributions are included in the performance analysis.

AB - The proposed algorithm identifies whether or not a test photo belongs to a set of co-located training images based on its spatial proximity to the training set. We leverage concepts from Lowe's SIFT and Snavely's Photo Tourism algorithms, and match an image by its 2D features to the 3D features representing the training set. To reduce complexity and increase efficiency, the proposed algorithm implements a compact representation of the image set by merging collections similar features. Test images are then matched with the derived structure. Finally, a decision statistic is determined based on the percentage of features that match. Receiver operating characteristics, computational analysis, and distributions are included in the performance analysis.

KW - 3D matching

KW - Approximate nearest neighbor

KW - SIFT

KW - Structure from motion

UR - http://www.scopus.com/inward/record.url?scp=78049405077&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78049405077&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2010.5495705

DO - 10.1109/ICASSP.2010.5495705

M3 - Conference contribution

AN - SCOPUS:78049405077

SN - 9781424442966

SP - 2194

EP - 2197

BT - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

ER -