Low Cost Video Animation of People using a RGBD Sensor

Cathrine Juel Thomsen, Thomas B. Moeslund, Troels Høg Peter Jensen

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

Abstract

This paper is an investigation in a low cost solution for performing video animation using a Kinect v2 for Windows, where skeleton, depth and colour data are acquired for three different characters. Segmentation of colour and depth frames were based on establishing the range of a person in the depth frame using the skeleton information, and then train a plane of the floor and exclude points close to it. Transitioning between motions were based on minimizing the Euclidean distance between all feasible transitioning frames, where a source and target frame would be found. Intermediate frames were made to create seamless transitions, where new poses were found by moving pixels in the direction of the optical flow between the transitioning frames. The realism of the proposed animation was verified through a user study to have a higher rate of preference and perceived realism compared to no animation and animation using alpha blending.
Original languageEnglish
Title of host publicationProceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 1: GRAPP
Volume1
PublisherSCITEPRESS Digital Library
Publication date2017
Pages244-249
ISBN (Print)978-989-758-224-0
DOIs
Publication statusPublished - 2017
EventThe 12th Internmational Joint Conference on Computer Vision, Imaging and Computer Graphics theory and Applications -
Duration: 1 Jan 2017 → …

Conference

ConferenceThe 12th Internmational Joint Conference on Computer Vision, Imaging and Computer Graphics theory and Applications
Period01/01/2017 → …

Fingerprint

Animation
Sensors
Costs
Color
Optical flows
Pixels

Cite this

Thomsen, C. J., Moeslund, T. B., & Jensen, T. H. P. (2017). Low Cost Video Animation of People using a RGBD Sensor. In Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 1: GRAPP (Vol. 1, pp. 244-249). SCITEPRESS Digital Library. https://doi.org/10.5220/0006132902440249
Thomsen, Cathrine Juel ; Moeslund, Thomas B. ; Jensen, Troels Høg Peter. / Low Cost Video Animation of People using a RGBD Sensor. Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 1: GRAPP. Vol. 1 SCITEPRESS Digital Library, 2017. pp. 244-249
@inproceedings{edfe0b81154e42ce97be3a0736b39d17,
title = "Low Cost Video Animation of People using a RGBD Sensor",
abstract = "This paper is an investigation in a low cost solution for performing video animation using a Kinect v2 for Windows, where skeleton, depth and colour data are acquired for three different characters. Segmentation of colour and depth frames were based on establishing the range of a person in the depth frame using the skeleton information, and then train a plane of the floor and exclude points close to it. Transitioning between motions were based on minimizing the Euclidean distance between all feasible transitioning frames, where a source and target frame would be found. Intermediate frames were made to create seamless transitions, where new poses were found by moving pixels in the direction of the optical flow between the transitioning frames. The realism of the proposed animation was verified through a user study to have a higher rate of preference and perceived realism compared to no animation and animation using alpha blending.",
author = "Thomsen, {Cathrine Juel} and Moeslund, {Thomas B.} and Jensen, {Troels H{\o}g Peter}",
year = "2017",
doi = "10.5220/0006132902440249",
language = "English",
isbn = "978-989-758-224-0",
volume = "1",
pages = "244--249",
booktitle = "Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 1: GRAPP",
publisher = "SCITEPRESS Digital Library",

}

Thomsen, CJ, Moeslund, TB & Jensen, THP 2017, Low Cost Video Animation of People using a RGBD Sensor. in Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 1: GRAPP. vol. 1, SCITEPRESS Digital Library, pp. 244-249, The 12th Internmational Joint Conference on Computer Vision, Imaging and Computer Graphics theory and Applications, 01/01/2017. https://doi.org/10.5220/0006132902440249

Low Cost Video Animation of People using a RGBD Sensor. / Thomsen, Cathrine Juel; Moeslund, Thomas B.; Jensen, Troels Høg Peter.

Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 1: GRAPP. Vol. 1 SCITEPRESS Digital Library, 2017. p. 244-249.

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

TY - GEN

T1 - Low Cost Video Animation of People using a RGBD Sensor

AU - Thomsen, Cathrine Juel

AU - Moeslund, Thomas B.

AU - Jensen, Troels Høg Peter

PY - 2017

Y1 - 2017

N2 - This paper is an investigation in a low cost solution for performing video animation using a Kinect v2 for Windows, where skeleton, depth and colour data are acquired for three different characters. Segmentation of colour and depth frames were based on establishing the range of a person in the depth frame using the skeleton information, and then train a plane of the floor and exclude points close to it. Transitioning between motions were based on minimizing the Euclidean distance between all feasible transitioning frames, where a source and target frame would be found. Intermediate frames were made to create seamless transitions, where new poses were found by moving pixels in the direction of the optical flow between the transitioning frames. The realism of the proposed animation was verified through a user study to have a higher rate of preference and perceived realism compared to no animation and animation using alpha blending.

AB - This paper is an investigation in a low cost solution for performing video animation using a Kinect v2 for Windows, where skeleton, depth and colour data are acquired for three different characters. Segmentation of colour and depth frames were based on establishing the range of a person in the depth frame using the skeleton information, and then train a plane of the floor and exclude points close to it. Transitioning between motions were based on minimizing the Euclidean distance between all feasible transitioning frames, where a source and target frame would be found. Intermediate frames were made to create seamless transitions, where new poses were found by moving pixels in the direction of the optical flow between the transitioning frames. The realism of the proposed animation was verified through a user study to have a higher rate of preference and perceived realism compared to no animation and animation using alpha blending.

U2 - 10.5220/0006132902440249

DO - 10.5220/0006132902440249

M3 - Article in proceeding

SN - 978-989-758-224-0

VL - 1

SP - 244

EP - 249

BT - Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 1: GRAPP

PB - SCITEPRESS Digital Library

ER -

Thomsen CJ, Moeslund TB, Jensen THP. Low Cost Video Animation of People using a RGBD Sensor. In Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 1: GRAPP. Vol. 1. SCITEPRESS Digital Library. 2017. p. 244-249 https://doi.org/10.5220/0006132902440249