Spatially and temporally optimized video stabilization

Yu Shuen Wang, Feng Liu, Pu Sheng Hsu, Tong Yee Lee

Research output: Contribution to journalArticle

66 Citations (Scopus)

Abstract

Properly handling parallax is important for video stabilization. Existing methods that achieve the aim require either 3D reconstruction or long feature trajectories to enforce the subspace or epipolar geometry constraints. In this paper, we present a robust and efficient technique that works on general videos. It achieves high-quality camera motion on videos where 3D reconstruction is difficult or long feature trajectories are not available. We represent each trajectory as a B ézier curve and maintain the spatial relations between trajectories by preserving the original offsets of neighboring curves. Our technique formulates stabilization as a spatial-temporal optimization problem that finds smooth feature trajectories and avoids visual distortion. The B ézier representation enables strong smoothness of each feature trajectory and reduces the number of variables in the optimization problem. We also stabilize videos in a streaming fashion to achieve scalability. The experiments show that our technique achieves high-quality camera motion on a variety of challenging videos that are difficult for existing methods.

Original languageEnglish
Article number6420828
Pages (from-to)1354-1361
Number of pages8
JournalIEEE Transactions on Visualization and Computer Graphics
Volume19
Issue number8
DOIs
Publication statusPublished - 2013 Jun 26

Fingerprint

Stabilization
Trajectories
Cameras
Scalability
Geometry
Experiments

All Science Journal Classification (ASJC) codes

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Computer Graphics and Computer-Aided Design

Cite this

Wang, Yu Shuen ; Liu, Feng ; Hsu, Pu Sheng ; Lee, Tong Yee. / Spatially and temporally optimized video stabilization. In: IEEE Transactions on Visualization and Computer Graphics. 2013 ; Vol. 19, No. 8. pp. 1354-1361.
@article{812dd539bd7c4a3b82105b8309616d01,
title = "Spatially and temporally optimized video stabilization",
abstract = "Properly handling parallax is important for video stabilization. Existing methods that achieve the aim require either 3D reconstruction or long feature trajectories to enforce the subspace or epipolar geometry constraints. In this paper, we present a robust and efficient technique that works on general videos. It achieves high-quality camera motion on videos where 3D reconstruction is difficult or long feature trajectories are not available. We represent each trajectory as a B {\'e}zier curve and maintain the spatial relations between trajectories by preserving the original offsets of neighboring curves. Our technique formulates stabilization as a spatial-temporal optimization problem that finds smooth feature trajectories and avoids visual distortion. The B {\'e}zier representation enables strong smoothness of each feature trajectory and reduces the number of variables in the optimization problem. We also stabilize videos in a streaming fashion to achieve scalability. The experiments show that our technique achieves high-quality camera motion on a variety of challenging videos that are difficult for existing methods.",
author = "Wang, {Yu Shuen} and Feng Liu and Hsu, {Pu Sheng} and Lee, {Tong Yee}",
year = "2013",
month = "6",
day = "26",
doi = "10.1109/TVCG.2013.11",
language = "English",
volume = "19",
pages = "1354--1361",
journal = "IEEE Transactions on Visualization and Computer Graphics",
issn = "1077-2626",
publisher = "IEEE Computer Society",
number = "8",

}

Spatially and temporally optimized video stabilization. / Wang, Yu Shuen; Liu, Feng; Hsu, Pu Sheng; Lee, Tong Yee.

In: IEEE Transactions on Visualization and Computer Graphics, Vol. 19, No. 8, 6420828, 26.06.2013, p. 1354-1361.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Spatially and temporally optimized video stabilization

AU - Wang, Yu Shuen

AU - Liu, Feng

AU - Hsu, Pu Sheng

AU - Lee, Tong Yee

PY - 2013/6/26

Y1 - 2013/6/26

N2 - Properly handling parallax is important for video stabilization. Existing methods that achieve the aim require either 3D reconstruction or long feature trajectories to enforce the subspace or epipolar geometry constraints. In this paper, we present a robust and efficient technique that works on general videos. It achieves high-quality camera motion on videos where 3D reconstruction is difficult or long feature trajectories are not available. We represent each trajectory as a B ézier curve and maintain the spatial relations between trajectories by preserving the original offsets of neighboring curves. Our technique formulates stabilization as a spatial-temporal optimization problem that finds smooth feature trajectories and avoids visual distortion. The B ézier representation enables strong smoothness of each feature trajectory and reduces the number of variables in the optimization problem. We also stabilize videos in a streaming fashion to achieve scalability. The experiments show that our technique achieves high-quality camera motion on a variety of challenging videos that are difficult for existing methods.

AB - Properly handling parallax is important for video stabilization. Existing methods that achieve the aim require either 3D reconstruction or long feature trajectories to enforce the subspace or epipolar geometry constraints. In this paper, we present a robust and efficient technique that works on general videos. It achieves high-quality camera motion on videos where 3D reconstruction is difficult or long feature trajectories are not available. We represent each trajectory as a B ézier curve and maintain the spatial relations between trajectories by preserving the original offsets of neighboring curves. Our technique formulates stabilization as a spatial-temporal optimization problem that finds smooth feature trajectories and avoids visual distortion. The B ézier representation enables strong smoothness of each feature trajectory and reduces the number of variables in the optimization problem. We also stabilize videos in a streaming fashion to achieve scalability. The experiments show that our technique achieves high-quality camera motion on a variety of challenging videos that are difficult for existing methods.

UR - http://www.scopus.com/inward/record.url?scp=84879236721&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84879236721&partnerID=8YFLogxK

U2 - 10.1109/TVCG.2013.11

DO - 10.1109/TVCG.2013.11

M3 - Article

C2 - 23744265

AN - SCOPUS:84879236721

VL - 19

SP - 1354

EP - 1361

JO - IEEE Transactions on Visualization and Computer Graphics

JF - IEEE Transactions on Visualization and Computer Graphics

SN - 1077-2626

IS - 8

M1 - 6420828

ER -