Live video object tracking and segmentation using graph cuts

Zachary Garrett, Hideo Saito

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

8 Citations (Scopus)

Abstract

Graph cuts have proven to be powerful tools in image segmentation. Previous graph cut research has proposed methods for cutting across large graphs constructed from multiple layered video frames, resulting in an object being tracked across multiple frames. However, this research focuses on cutting graphs constructed from a prerecorded video sequence. In live video scenarios, frames cannot be layered to construct 3D volumes, since the contents of the subsequent frames are unknown. Instead, new graphs must be created and cut for each frame on demand. Resource limitations make this unfeasible on high-resolution videos. In addition, object tracking requires a method for incorporating the previous frame's object position and shape into the current graph. We propose a method for tracking and segmenting objects in live video that utilizes regional graph cuts and object pixel probability maps. The regionalization of the cuts around the tracked object will increase the speed of the tracker, and the object pixel probability maps will enable more flexible tracking.

Original languageEnglish
Title of host publicationProceedings - International Conference on Image Processing, ICIP
Pages1576-1579
Number of pages4
DOIs
Publication statusPublished - 2008
Event2008 IEEE International Conference on Image Processing, ICIP 2008 - San Diego, CA, United States
Duration: 2008 Oct 122008 Oct 15

Other

Other2008 IEEE International Conference on Image Processing, ICIP 2008
CountryUnited States
CitySan Diego, CA
Period08/10/1208/10/15

Fingerprint

Pixels
Image segmentation

Keywords

  • Image processing
  • Image segmentation
  • Real time systems
  • Tracking

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition
  • Signal Processing

Cite this

Garrett, Z., & Saito, H. (2008). Live video object tracking and segmentation using graph cuts. In Proceedings - International Conference on Image Processing, ICIP (pp. 1576-1579). [4712070] https://doi.org/10.1109/ICIP.2008.4712070

Live video object tracking and segmentation using graph cuts. / Garrett, Zachary; Saito, Hideo.

Proceedings - International Conference on Image Processing, ICIP. 2008. p. 1576-1579 4712070.

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

Garrett, Z & Saito, H 2008, Live video object tracking and segmentation using graph cuts. in Proceedings - International Conference on Image Processing, ICIP., 4712070, pp. 1576-1579, 2008 IEEE International Conference on Image Processing, ICIP 2008, San Diego, CA, United States, 08/10/12. https://doi.org/10.1109/ICIP.2008.4712070
Garrett Z, Saito H. Live video object tracking and segmentation using graph cuts. In Proceedings - International Conference on Image Processing, ICIP. 2008. p. 1576-1579. 4712070 https://doi.org/10.1109/ICIP.2008.4712070
Garrett, Zachary ; Saito, Hideo. / Live video object tracking and segmentation using graph cuts. Proceedings - International Conference on Image Processing, ICIP. 2008. pp. 1576-1579
@inproceedings{70c58a36c0c749ca9dfd0a97d1190987,
title = "Live video object tracking and segmentation using graph cuts",
abstract = "Graph cuts have proven to be powerful tools in image segmentation. Previous graph cut research has proposed methods for cutting across large graphs constructed from multiple layered video frames, resulting in an object being tracked across multiple frames. However, this research focuses on cutting graphs constructed from a prerecorded video sequence. In live video scenarios, frames cannot be layered to construct 3D volumes, since the contents of the subsequent frames are unknown. Instead, new graphs must be created and cut for each frame on demand. Resource limitations make this unfeasible on high-resolution videos. In addition, object tracking requires a method for incorporating the previous frame's object position and shape into the current graph. We propose a method for tracking and segmenting objects in live video that utilizes regional graph cuts and object pixel probability maps. The regionalization of the cuts around the tracked object will increase the speed of the tracker, and the object pixel probability maps will enable more flexible tracking.",
keywords = "Image processing, Image segmentation, Real time systems, Tracking",
author = "Zachary Garrett and Hideo Saito",
year = "2008",
doi = "10.1109/ICIP.2008.4712070",
language = "English",
isbn = "1424417643",
pages = "1576--1579",
booktitle = "Proceedings - International Conference on Image Processing, ICIP",

}

TY - GEN

T1 - Live video object tracking and segmentation using graph cuts

AU - Garrett, Zachary

AU - Saito, Hideo

PY - 2008

Y1 - 2008

N2 - Graph cuts have proven to be powerful tools in image segmentation. Previous graph cut research has proposed methods for cutting across large graphs constructed from multiple layered video frames, resulting in an object being tracked across multiple frames. However, this research focuses on cutting graphs constructed from a prerecorded video sequence. In live video scenarios, frames cannot be layered to construct 3D volumes, since the contents of the subsequent frames are unknown. Instead, new graphs must be created and cut for each frame on demand. Resource limitations make this unfeasible on high-resolution videos. In addition, object tracking requires a method for incorporating the previous frame's object position and shape into the current graph. We propose a method for tracking and segmenting objects in live video that utilizes regional graph cuts and object pixel probability maps. The regionalization of the cuts around the tracked object will increase the speed of the tracker, and the object pixel probability maps will enable more flexible tracking.

AB - Graph cuts have proven to be powerful tools in image segmentation. Previous graph cut research has proposed methods for cutting across large graphs constructed from multiple layered video frames, resulting in an object being tracked across multiple frames. However, this research focuses on cutting graphs constructed from a prerecorded video sequence. In live video scenarios, frames cannot be layered to construct 3D volumes, since the contents of the subsequent frames are unknown. Instead, new graphs must be created and cut for each frame on demand. Resource limitations make this unfeasible on high-resolution videos. In addition, object tracking requires a method for incorporating the previous frame's object position and shape into the current graph. We propose a method for tracking and segmenting objects in live video that utilizes regional graph cuts and object pixel probability maps. The regionalization of the cuts around the tracked object will increase the speed of the tracker, and the object pixel probability maps will enable more flexible tracking.

KW - Image processing

KW - Image segmentation

KW - Real time systems

KW - Tracking

UR - http://www.scopus.com/inward/record.url?scp=69949123418&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=69949123418&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2008.4712070

DO - 10.1109/ICIP.2008.4712070

M3 - Conference contribution

AN - SCOPUS:69949123418

SN - 1424417643

SN - 9781424417643

SP - 1576

EP - 1579

BT - Proceedings - International Conference on Image Processing, ICIP

ER -