Motion-based video retargeting with optimized crop-and-warp

Yu Shuen Wang, Hui Chih Lin, Olga Sorkine, Tong Yee Lee

Research output: Contribution to journalArticle

78 Citations (Scopus)

Abstract

We introduce a video retargeting method that achieves high-quality resizing to arbitrary aspect ratios for complex videos containing diverse camera and dynamic motions. Previous content-aware retargeting methods mostly concentrated on spatial considerations, attempting to preserve the shape of salient objects in each frame by removing or distorting homogeneous background content. However, sacrificeable space is fundamentally limited in video, since object motion makes foreground and background regions correlated, causing waving and squeezing artifacts. We solve the retargeting problem by explicitly employing motion information and by distributing distortion in both spatial and temporal dimensions. We combine novel cropping and warping operators, where the cropping removes temporally-recurring contents and the warping utilizes available homogeneous regions to mask deformations while preserving motion. Variational optimization allows to find the best balance between the two operations, enabling retargeting of challenging videos with complex motions, numerous prominent objects and arbitrary depth variability. Our method compares favorably with state-of-the-art retargeting systems, as demonstrated in the examples and widely supported by the conducted user study.

Original languageEnglish
Article number90
JournalACM Transactions on Graphics
Volume29
Issue number4
DOIs
Publication statusPublished - 2010 Sep 13

Fingerprint

Crops
Aspect ratio
Masks
Cameras

All Science Journal Classification (ASJC) codes

  • Computer Graphics and Computer-Aided Design

Cite this

Wang, Yu Shuen ; Lin, Hui Chih ; Sorkine, Olga ; Lee, Tong Yee. / Motion-based video retargeting with optimized crop-and-warp. In: ACM Transactions on Graphics. 2010 ; Vol. 29, No. 4.
@article{6c384cbe12de4b7792981f2af741c7f2,
title = "Motion-based video retargeting with optimized crop-and-warp",
abstract = "We introduce a video retargeting method that achieves high-quality resizing to arbitrary aspect ratios for complex videos containing diverse camera and dynamic motions. Previous content-aware retargeting methods mostly concentrated on spatial considerations, attempting to preserve the shape of salient objects in each frame by removing or distorting homogeneous background content. However, sacrificeable space is fundamentally limited in video, since object motion makes foreground and background regions correlated, causing waving and squeezing artifacts. We solve the retargeting problem by explicitly employing motion information and by distributing distortion in both spatial and temporal dimensions. We combine novel cropping and warping operators, where the cropping removes temporally-recurring contents and the warping utilizes available homogeneous regions to mask deformations while preserving motion. Variational optimization allows to find the best balance between the two operations, enabling retargeting of challenging videos with complex motions, numerous prominent objects and arbitrary depth variability. Our method compares favorably with state-of-the-art retargeting systems, as demonstrated in the examples and widely supported by the conducted user study.",
author = "Wang, {Yu Shuen} and Lin, {Hui Chih} and Olga Sorkine and Lee, {Tong Yee}",
year = "2010",
month = "9",
day = "13",
doi = "10.1145/1778765.1778827",
language = "English",
volume = "29",
journal = "ACM Transactions on Graphics",
issn = "0730-0301",
publisher = "Association for Computing Machinery (ACM)",
number = "4",

}

Motion-based video retargeting with optimized crop-and-warp. / Wang, Yu Shuen; Lin, Hui Chih; Sorkine, Olga; Lee, Tong Yee.

In: ACM Transactions on Graphics, Vol. 29, No. 4, 90, 13.09.2010.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Motion-based video retargeting with optimized crop-and-warp

AU - Wang, Yu Shuen

AU - Lin, Hui Chih

AU - Sorkine, Olga

AU - Lee, Tong Yee

PY - 2010/9/13

Y1 - 2010/9/13

N2 - We introduce a video retargeting method that achieves high-quality resizing to arbitrary aspect ratios for complex videos containing diverse camera and dynamic motions. Previous content-aware retargeting methods mostly concentrated on spatial considerations, attempting to preserve the shape of salient objects in each frame by removing or distorting homogeneous background content. However, sacrificeable space is fundamentally limited in video, since object motion makes foreground and background regions correlated, causing waving and squeezing artifacts. We solve the retargeting problem by explicitly employing motion information and by distributing distortion in both spatial and temporal dimensions. We combine novel cropping and warping operators, where the cropping removes temporally-recurring contents and the warping utilizes available homogeneous regions to mask deformations while preserving motion. Variational optimization allows to find the best balance between the two operations, enabling retargeting of challenging videos with complex motions, numerous prominent objects and arbitrary depth variability. Our method compares favorably with state-of-the-art retargeting systems, as demonstrated in the examples and widely supported by the conducted user study.

AB - We introduce a video retargeting method that achieves high-quality resizing to arbitrary aspect ratios for complex videos containing diverse camera and dynamic motions. Previous content-aware retargeting methods mostly concentrated on spatial considerations, attempting to preserve the shape of salient objects in each frame by removing or distorting homogeneous background content. However, sacrificeable space is fundamentally limited in video, since object motion makes foreground and background regions correlated, causing waving and squeezing artifacts. We solve the retargeting problem by explicitly employing motion information and by distributing distortion in both spatial and temporal dimensions. We combine novel cropping and warping operators, where the cropping removes temporally-recurring contents and the warping utilizes available homogeneous regions to mask deformations while preserving motion. Variational optimization allows to find the best balance between the two operations, enabling retargeting of challenging videos with complex motions, numerous prominent objects and arbitrary depth variability. Our method compares favorably with state-of-the-art retargeting systems, as demonstrated in the examples and widely supported by the conducted user study.

UR - http://www.scopus.com/inward/record.url?scp=77956376502&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77956376502&partnerID=8YFLogxK

U2 - 10.1145/1778765.1778827

DO - 10.1145/1778765.1778827

M3 - Article

AN - SCOPUS:77956376502

VL - 29

JO - ACM Transactions on Graphics

JF - ACM Transactions on Graphics

SN - 0730-0301

IS - 4

M1 - 90

ER -