An unsupervised multi-resolution object extraction algorithm using video-cube

F. M. Porikli, Yao Wang

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

Abstract

We propose a fast video object segmentation method that detects object boundaries accurately, and does not require any user assistance. Video streams are considered as 3D data, called video-cubes, to take advantage of 3D signal processing techniques. After a video sequence is filtered, marker nodes are selected from the color gradient. A volume around each marker is grown by using color/texture distance criteria. Then volumes that have similar characteristics are merged. Self-descriptors for each volume, mutual-descriptors for each pair of volumes are computed. These descriptors capture motion and spatial information of volumes. In the clustering stage, volumes are classified into objects in a fine-to-coarse hierarchy. While applying and relaxing descriptor based adaptive, similarity scores are estimated for each possible pair-wise combination of volumes. The pair that gives the maximum score is clustered iteratively. Finally, an object-based multi-resolution representation tree is assembled.

Original languageEnglish (US)
Title of host publicationIEEE International Conference on Image Processing
Pages359-362
Number of pages4
Volume2
StatePublished - 2001
EventIEEE International Conference on Image Processing (ICIP) - Thessaloniki, Greece
Duration: Oct 7 2001Oct 10 2001

Other

OtherIEEE International Conference on Image Processing (ICIP)
CountryGreece
CityThessaloniki
Period10/7/0110/10/01

Fingerprint

Color
Signal processing
Textures

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Hardware and Architecture
  • Electrical and Electronic Engineering

Cite this

Porikli, F. M., & Wang, Y. (2001). An unsupervised multi-resolution object extraction algorithm using video-cube. In IEEE International Conference on Image Processing (Vol. 2, pp. 359-362)

An unsupervised multi-resolution object extraction algorithm using video-cube. / Porikli, F. M.; Wang, Yao.

IEEE International Conference on Image Processing. Vol. 2 2001. p. 359-362.

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

Porikli, FM & Wang, Y 2001, An unsupervised multi-resolution object extraction algorithm using video-cube. in IEEE International Conference on Image Processing. vol. 2, pp. 359-362, IEEE International Conference on Image Processing (ICIP), Thessaloniki, Greece, 10/7/01.
Porikli FM, Wang Y. An unsupervised multi-resolution object extraction algorithm using video-cube. In IEEE International Conference on Image Processing. Vol. 2. 2001. p. 359-362
Porikli, F. M. ; Wang, Yao. / An unsupervised multi-resolution object extraction algorithm using video-cube. IEEE International Conference on Image Processing. Vol. 2 2001. pp. 359-362
@inproceedings{3b420657d02a4307af5daeb656e15d6b,
title = "An unsupervised multi-resolution object extraction algorithm using video-cube",
abstract = "We propose a fast video object segmentation method that detects object boundaries accurately, and does not require any user assistance. Video streams are considered as 3D data, called video-cubes, to take advantage of 3D signal processing techniques. After a video sequence is filtered, marker nodes are selected from the color gradient. A volume around each marker is grown by using color/texture distance criteria. Then volumes that have similar characteristics are merged. Self-descriptors for each volume, mutual-descriptors for each pair of volumes are computed. These descriptors capture motion and spatial information of volumes. In the clustering stage, volumes are classified into objects in a fine-to-coarse hierarchy. While applying and relaxing descriptor based adaptive, similarity scores are estimated for each possible pair-wise combination of volumes. The pair that gives the maximum score is clustered iteratively. Finally, an object-based multi-resolution representation tree is assembled.",
author = "Porikli, {F. M.} and Yao Wang",
year = "2001",
language = "English (US)",
volume = "2",
pages = "359--362",
booktitle = "IEEE International Conference on Image Processing",

}

TY - GEN

T1 - An unsupervised multi-resolution object extraction algorithm using video-cube

AU - Porikli, F. M.

AU - Wang, Yao

PY - 2001

Y1 - 2001

N2 - We propose a fast video object segmentation method that detects object boundaries accurately, and does not require any user assistance. Video streams are considered as 3D data, called video-cubes, to take advantage of 3D signal processing techniques. After a video sequence is filtered, marker nodes are selected from the color gradient. A volume around each marker is grown by using color/texture distance criteria. Then volumes that have similar characteristics are merged. Self-descriptors for each volume, mutual-descriptors for each pair of volumes are computed. These descriptors capture motion and spatial information of volumes. In the clustering stage, volumes are classified into objects in a fine-to-coarse hierarchy. While applying and relaxing descriptor based adaptive, similarity scores are estimated for each possible pair-wise combination of volumes. The pair that gives the maximum score is clustered iteratively. Finally, an object-based multi-resolution representation tree is assembled.

AB - We propose a fast video object segmentation method that detects object boundaries accurately, and does not require any user assistance. Video streams are considered as 3D data, called video-cubes, to take advantage of 3D signal processing techniques. After a video sequence is filtered, marker nodes are selected from the color gradient. A volume around each marker is grown by using color/texture distance criteria. Then volumes that have similar characteristics are merged. Self-descriptors for each volume, mutual-descriptors for each pair of volumes are computed. These descriptors capture motion and spatial information of volumes. In the clustering stage, volumes are classified into objects in a fine-to-coarse hierarchy. While applying and relaxing descriptor based adaptive, similarity scores are estimated for each possible pair-wise combination of volumes. The pair that gives the maximum score is clustered iteratively. Finally, an object-based multi-resolution representation tree is assembled.

UR - http://www.scopus.com/inward/record.url?scp=0035167304&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0035167304&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0035167304

VL - 2

SP - 359

EP - 362

BT - IEEE International Conference on Image Processing

ER -