Adaptive multiscale optical flow estimation

J Li; CP Benton; SG Nikolov; NE Scott-Samuel

doi:10.1109/ICIP.2007.4379204

Adaptive multiscale optical flow estimation

J Li, CP Benton, SG Nikolov, NE Scott-Samuel

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

Abstract

The current paper presents a novel adaptive multiscale scheme to estimate optical flow from image sequences. The scheme models estimation uncertainties which are used to reduce the influence of unreliable intermediate estimates on accuracy. The experimental results show that the proposed method provides more accurate estimates for both small and large motions than a standard multiscale scheme in which an increment is added to an intermediate estimate regardless of estimation certainty.

Translated title of the contribution	Adaptive multiscale optical flow estimation
Original language	English
Title of host publication	IEEE Conference on Image Processing, San Antonio, TX
Pages	1073 - 1076
Number of pages	4
Volume	1-7
DOIs	https://doi.org/10.1109/ICIP.2007.4379204
Publication status	Published - 16 Sept 2007

Access to Document

10.1109/ICIP.2007.4379204

Handle.net

Persistent link

Cite this

@inproceedings{2f42a540e1dd4063933f259f01885fc1,

title = "Adaptive multiscale optical flow estimation",

abstract = "The current paper presents a novel adaptive multiscale scheme to estimate optical flow from image sequences. The scheme models estimation uncertainties which are used to reduce the influence of unreliable intermediate estimates on accuracy. The experimental results show that the proposed method provides more accurate estimates for both small and large motions than a standard multiscale scheme in which an increment is added to an intermediate estimate regardless of estimation certainty.",

author = "J Li and CP Benton and SG Nikolov and NE Scott-Samuel",

year = "2007",

month = sep,

day = "16",

doi = "10.1109/ICIP.2007.4379204",

language = "English",

isbn = "9781424414376",

volume = "1-7",

pages = "1073 -- 1076",

booktitle = "IEEE Conference on Image Processing, San Antonio, TX",

}

TY - GEN

T1 - Adaptive multiscale optical flow estimation

AU - Li, J

AU - Benton, CP

AU - Nikolov, SG

AU - Scott-Samuel, NE

PY - 2007/9/16

Y1 - 2007/9/16

N2 - The current paper presents a novel adaptive multiscale scheme to estimate optical flow from image sequences. The scheme models estimation uncertainties which are used to reduce the influence of unreliable intermediate estimates on accuracy. The experimental results show that the proposed method provides more accurate estimates for both small and large motions than a standard multiscale scheme in which an increment is added to an intermediate estimate regardless of estimation certainty.

AB - The current paper presents a novel adaptive multiscale scheme to estimate optical flow from image sequences. The scheme models estimation uncertainties which are used to reduce the influence of unreliable intermediate estimates on accuracy. The experimental results show that the proposed method provides more accurate estimates for both small and large motions than a standard multiscale scheme in which an increment is added to an intermediate estimate regardless of estimation certainty.

U2 - 10.1109/ICIP.2007.4379204

DO - 10.1109/ICIP.2007.4379204

M3 - Conference Contribution (Conference Proceeding)

SN - 9781424414376

VL - 1-7

SP - 1073

EP - 1076

BT - IEEE Conference on Image Processing, San Antonio, TX

ER -

Adaptive multiscale optical flow estimation

Abstract

Access to Document

Handle.net

Fingerprint

Cite this