Real-time video mosaicking robust to dynamic scenes

Nan Jiang, Glen Abousleman, Jennie Si

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

Abstract

This paper presents a system for creating a mosaic image from a sequence of images with moving objects present in the scene. This system first uses SIFT-based image registration on the entire image to obtain the initial global projection matrix. After image segmentation, the global motion model is applied to each region for evaluation. The transformation matrix is refined for best projection in each region, and a more precise global transformation matrix is calculated based upon local projections on majority of coherent regions. As a consequence this method is robust to disturbances to the projection model induced by moving objects and motion parallax. In the image blending stage, pixels in coherent regions are weighted by their distances from the overlapping edges to achieve a seamless panorama, while heterogeneous regions are cut and pasted to avoid ghosting or blurring. The most recent information regarding location, shape, and size of the moving foreground objects is therefore reflected in the panorama. Constructed mosaics are presented to demonstrate the performance and robustness of the proposed algorithm.

Original languageEnglish (US)
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
Volume6209
DOIs
StatePublished - 2006
EventAirborne Intelligence, Surveillance, Reconnaissance (ISR) Systems and Applications III - Kissimmee, FL, United States
Duration: Apr 19 2006Apr 20 2006

Other

OtherAirborne Intelligence, Surveillance, Reconnaissance (ISR) Systems and Applications III
CountryUnited States
CityKissimmee, FL
Period4/19/064/20/06

Fingerprint

projection
Image registration
Image segmentation
Pixels
blurring
parallax
disturbances
pixels
evaluation
matrices

Keywords

  • Feature points
  • Image registration
  • Video mosaic

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Condensed Matter Physics

Cite this

Jiang, N., Abousleman, G., & Si, J. (2006). Real-time video mosaicking robust to dynamic scenes. In Proceedings of SPIE - The International Society for Optical Engineering (Vol. 6209). [620907] https://doi.org/10.1117/12.665880

Real-time video mosaicking robust to dynamic scenes. / Jiang, Nan; Abousleman, Glen; Si, Jennie.

Proceedings of SPIE - The International Society for Optical Engineering. Vol. 6209 2006. 620907.

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

Jiang, N, Abousleman, G & Si, J 2006, Real-time video mosaicking robust to dynamic scenes. in Proceedings of SPIE - The International Society for Optical Engineering. vol. 6209, 620907, Airborne Intelligence, Surveillance, Reconnaissance (ISR) Systems and Applications III, Kissimmee, FL, United States, 4/19/06. https://doi.org/10.1117/12.665880
Jiang N, Abousleman G, Si J. Real-time video mosaicking robust to dynamic scenes. In Proceedings of SPIE - The International Society for Optical Engineering. Vol. 6209. 2006. 620907 https://doi.org/10.1117/12.665880
Jiang, Nan ; Abousleman, Glen ; Si, Jennie. / Real-time video mosaicking robust to dynamic scenes. Proceedings of SPIE - The International Society for Optical Engineering. Vol. 6209 2006.
@inproceedings{f911109b96c8484c9f3a260399c9d9aa,
title = "Real-time video mosaicking robust to dynamic scenes",
abstract = "This paper presents a system for creating a mosaic image from a sequence of images with moving objects present in the scene. This system first uses SIFT-based image registration on the entire image to obtain the initial global projection matrix. After image segmentation, the global motion model is applied to each region for evaluation. The transformation matrix is refined for best projection in each region, and a more precise global transformation matrix is calculated based upon local projections on majority of coherent regions. As a consequence this method is robust to disturbances to the projection model induced by moving objects and motion parallax. In the image blending stage, pixels in coherent regions are weighted by their distances from the overlapping edges to achieve a seamless panorama, while heterogeneous regions are cut and pasted to avoid ghosting or blurring. The most recent information regarding location, shape, and size of the moving foreground objects is therefore reflected in the panorama. Constructed mosaics are presented to demonstrate the performance and robustness of the proposed algorithm.",
keywords = "Feature points, Image registration, Video mosaic",
author = "Nan Jiang and Glen Abousleman and Jennie Si",
year = "2006",
doi = "10.1117/12.665880",
language = "English (US)",
isbn = "0819462659",
volume = "6209",
booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",

}

TY - GEN

T1 - Real-time video mosaicking robust to dynamic scenes

AU - Jiang, Nan

AU - Abousleman, Glen

AU - Si, Jennie

PY - 2006

Y1 - 2006

N2 - This paper presents a system for creating a mosaic image from a sequence of images with moving objects present in the scene. This system first uses SIFT-based image registration on the entire image to obtain the initial global projection matrix. After image segmentation, the global motion model is applied to each region for evaluation. The transformation matrix is refined for best projection in each region, and a more precise global transformation matrix is calculated based upon local projections on majority of coherent regions. As a consequence this method is robust to disturbances to the projection model induced by moving objects and motion parallax. In the image blending stage, pixels in coherent regions are weighted by their distances from the overlapping edges to achieve a seamless panorama, while heterogeneous regions are cut and pasted to avoid ghosting or blurring. The most recent information regarding location, shape, and size of the moving foreground objects is therefore reflected in the panorama. Constructed mosaics are presented to demonstrate the performance and robustness of the proposed algorithm.

AB - This paper presents a system for creating a mosaic image from a sequence of images with moving objects present in the scene. This system first uses SIFT-based image registration on the entire image to obtain the initial global projection matrix. After image segmentation, the global motion model is applied to each region for evaluation. The transformation matrix is refined for best projection in each region, and a more precise global transformation matrix is calculated based upon local projections on majority of coherent regions. As a consequence this method is robust to disturbances to the projection model induced by moving objects and motion parallax. In the image blending stage, pixels in coherent regions are weighted by their distances from the overlapping edges to achieve a seamless panorama, while heterogeneous regions are cut and pasted to avoid ghosting or blurring. The most recent information regarding location, shape, and size of the moving foreground objects is therefore reflected in the panorama. Constructed mosaics are presented to demonstrate the performance and robustness of the proposed algorithm.

KW - Feature points

KW - Image registration

KW - Video mosaic

UR - http://www.scopus.com/inward/record.url?scp=33747743064&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33747743064&partnerID=8YFLogxK

U2 - 10.1117/12.665880

DO - 10.1117/12.665880

M3 - Conference contribution

AN - SCOPUS:33747743064

SN - 0819462659

SN - 9780819462657

VL - 6209

BT - Proceedings of SPIE - The International Society for Optical Engineering

ER -