The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
La vidéo émergente à trois degrés de liberté plus (3DoF+) offre une expérience visuelle immersive plus interactive et plus profonde. La vidéo 3DoF+ introduit la parallaxe de mouvement dans la vidéo 360°, offrant une vue omnidirectionnelle avec des changements limités de position de vue. Un large éventail de vues est nécessaire pour prendre en charge une telle expérience visuelle 3DoF+. Il est donc essentiel de compresser une énorme quantité de vidéo 3DoF+. Récemment, MPEG développe une norme pour un codage efficace de la vidéo 3DoF+ composée de plusieurs vidéos, ainsi que son modèle de test nommé Test Model for Immersive Video (TMIV). Dans le TMIV, la redondance entre les vues sources d'entrée est supprimée autant que possible en sélectionnant une ou plusieurs vues de base et en prédisant les vues restantes à partir des vues de base. Chaque région imprévue est recadrée dans un cadre de délimitation appelé patch, puis un grand nombre de patchs sont regroupés dans des atlas avec les vues de base sélectionnées. En conséquence, plusieurs vues sources sont converties en une ou plusieurs séquences d'atlas à compresser. Dans cette lettre, nous présentons une méthode de clustering améliorée utilisant la fusion de patchs dans la construction d'atlas dans le TMIV. La méthode proposée permet d'obtenir une réduction significative du taux de BD en termes de diverses mesures d'évaluation de bout en bout dans l'expérience et a été adoptée dans TMIV6.0.
Hyun-Ho KIM
Korea Aerospace University
Sung-Gyun LIM
Korea Aerospace University
Gwangsoon LEE
ETRI
Jun Young JEONG
ETRI
Jae-Gon KIM
Korea Aerospace University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copier
Hyun-Ho KIM, Sung-Gyun LIM, Gwangsoon LEE, Jun Young JEONG, Jae-Gon KIM, "Efficient Patch Merging for Atlas Construction in 3DoF+ Video Coding" in IEICE TRANSACTIONS on Information,
vol. E104-D, no. 3, pp. 477-480, March 2021, doi: 10.1587/transinf.2020EDL8119.
Abstract: The emerging three degree of freedom plus (3DoF+) video provides more interactive and deep immersive visual experience. 3DoF+ video introduces motion parallax to 360 video providing omnidirectional view with limited changes of the view position. A large set of views are required to support such 3DoF+ visual experience, hence it is essential to compress a tremendous amount of 3DoF+ video. Recently, MPEG is developing a standard for efficient coding of 3DoF+ video that consists of multiple videos, and its test model named Test Model for Immersive Video (TMIV). In the TMIV, the redundancy between the input source views is removed as much as possible by selecting one or several basic views and predicting the remaining views from the basic views. Each unpredicted region is cropped to a bounding box called patch, and then a large number of patches are packed into atlases together with the selected basic views. As a result, multiple source views are converted into one or more atlas sequences to be compressed. In this letter, we present an improved clustering method using patch merging in the atlas construction in the TMIV. The proposed method achieves significant BD-rate reduction in terms of various end-to-end evaluation metrics in the experiment, and was adopted in TMIV6.0.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2020EDL8119/_p
Copier
@ARTICLE{e104-d_3_477,
author={Hyun-Ho KIM, Sung-Gyun LIM, Gwangsoon LEE, Jun Young JEONG, Jae-Gon KIM, },
journal={IEICE TRANSACTIONS on Information},
title={Efficient Patch Merging for Atlas Construction in 3DoF+ Video Coding},
year={2021},
volume={E104-D},
number={3},
pages={477-480},
abstract={The emerging three degree of freedom plus (3DoF+) video provides more interactive and deep immersive visual experience. 3DoF+ video introduces motion parallax to 360 video providing omnidirectional view with limited changes of the view position. A large set of views are required to support such 3DoF+ visual experience, hence it is essential to compress a tremendous amount of 3DoF+ video. Recently, MPEG is developing a standard for efficient coding of 3DoF+ video that consists of multiple videos, and its test model named Test Model for Immersive Video (TMIV). In the TMIV, the redundancy between the input source views is removed as much as possible by selecting one or several basic views and predicting the remaining views from the basic views. Each unpredicted region is cropped to a bounding box called patch, and then a large number of patches are packed into atlases together with the selected basic views. As a result, multiple source views are converted into one or more atlas sequences to be compressed. In this letter, we present an improved clustering method using patch merging in the atlas construction in the TMIV. The proposed method achieves significant BD-rate reduction in terms of various end-to-end evaluation metrics in the experiment, and was adopted in TMIV6.0.},
keywords={},
doi={10.1587/transinf.2020EDL8119},
ISSN={1745-1361},
month={March},}
Copier
TY - JOUR
TI - Efficient Patch Merging for Atlas Construction in 3DoF+ Video Coding
T2 - IEICE TRANSACTIONS on Information
SP - 477
EP - 480
AU - Hyun-Ho KIM
AU - Sung-Gyun LIM
AU - Gwangsoon LEE
AU - Jun Young JEONG
AU - Jae-Gon KIM
PY - 2021
DO - 10.1587/transinf.2020EDL8119
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E104-D
IS - 3
JA - IEICE TRANSACTIONS on Information
Y1 - March 2021
AB - The emerging three degree of freedom plus (3DoF+) video provides more interactive and deep immersive visual experience. 3DoF+ video introduces motion parallax to 360 video providing omnidirectional view with limited changes of the view position. A large set of views are required to support such 3DoF+ visual experience, hence it is essential to compress a tremendous amount of 3DoF+ video. Recently, MPEG is developing a standard for efficient coding of 3DoF+ video that consists of multiple videos, and its test model named Test Model for Immersive Video (TMIV). In the TMIV, the redundancy between the input source views is removed as much as possible by selecting one or several basic views and predicting the remaining views from the basic views. Each unpredicted region is cropped to a bounding box called patch, and then a large number of patches are packed into atlases together with the selected basic views. As a result, multiple source views are converted into one or more atlas sequences to be compressed. In this letter, we present an improved clustering method using patch merging in the atlas construction in the TMIV. The proposed method achieves significant BD-rate reduction in terms of various end-to-end evaluation metrics in the experiment, and was adopted in TMIV6.0.
ER -