Improved FGS scalable scheme for multi-view stereo video

doi:10.3778/j.issn.1002-8331.2009.05.002

Computer Engineering and Applications ›› 2009, Vol. 45 ›› Issue (5): 7-10.DOI: 10.3778/j.issn.1002-8331.2009.05.002

• 博士论坛 • Previous Articles Next Articles

Improved FGS scalable scheme for multi-view stereo video

YANG Lei^1,2,SONG Xiao-wei²,HOU Chun-ping¹

1.School of Electronic and Information Engineering，Tianjin University，Tianjin 300072，China
2.School of Electronic and Information Engineering，Zhongyuan University of Technology，Zhengzhou 450007，China

Received:2008-09-25 Revised:2008-11-17 Online:2009-02-11 Published:2009-02-11
Contact: YANG Lei

改进的多视点立体视频FGS可分级方案

杨蕾^1,2,宋晓炜²,侯春萍¹

1.天津大学电子信息工程学院，天津 300072
2.中原工学院电子信息学院，郑州 450007

通讯作者: 杨蕾

Abstract

Abstract: To improve the low coding efficiency problem of an existing multi-view stereo video FGS scalable scheme，this paper makes full use of the similarity between adjacent views，adapts different demands on Quality-First（QF） or View-First（VF） of the decoder，and provides a tradeoff between image quality and view number.Improvement have been made for three situations，I，P，B frames，respectively.The mid-view is encoded as base layer，while other views are predicted from the partly retrieved FGS enhancement layers for adjacent views，thus produces the FGS enhancement layers for current view.Experimental results have shown that this scheme has more flexible and extensive scalable characteristic，which can adapt different demands on image quality and stereo impression of different users.

Key words: Fine-Granular-Scalability（FGS）, scalable coding, multi-view, stereo video

摘要： 针对现有多视点立体视频FGS（Fine-Granular-Scalability）可分级方案编码效率低的问题进行改进，充分利用相邻视点之间的相似性，适应解码端质量优先或视点优先的不同需求，提出一种折衷方案，其中分I、P、B帧三种情况，以中间视点作为基本层进行编码，其他视点均预测自相邻视点部分恢复的FGS增强层，以此为基础产生当前视点的FGS增强层。实验证明，该方案具有更为灵活、广泛的可分级性能，更能适应用户对视点图像质量和立体观感的需求。

关键词: 精细可分级视频编码（FGS）, 可分级, 多视点, 立体视频

YANG Lei^1,2,SONG Xiao-wei²,HOU Chun-ping¹. Improved FGS scalable scheme for multi-view stereo video[J]. Computer Engineering and Applications, 2009, 45(5): 7-10.

杨蕾^1,2,宋晓炜²,侯春萍¹. 改进的多视点立体视频FGS可分级方案[J]. 计算机工程与应用, 2009, 45(5): 7-10.

[1]	LIANG Fangxuan, YANG Feng, LU Liyun, YIN Mengxiao. Review of Brain Tumor Segmentation Methods Based on Convolutional Neural Networks [J]. Computer Engineering and Applications, 2021, 57(7): 34-43.
[2]	LI Yongzhen, LIAO Husheng. Multi-view Clustering via Graph Convolutional Neural Network [J]. Computer Engineering and Applications, 2021, 57(5): 115-122.
[3]	TANG Ning, WEI Zeliang, ZHANG Rui, YI Dong, WU Yazhou. Comparative Study on Classification of Pulmonary Nodules Based on Multi-Scale and Multi-Mode Image [J]. Computer Engineering and Applications, 2020, 56(3): 165-175.
[4]	GUO Sheng, ZHONG Zhaoman, LI Cunhua. Multi-view Subspace Clustering Network Based on Deep Autoencoder [J]. Computer Engineering and Applications, 2020, 56(17): 60-68.
[5]	NIU Dandan1, DUAN Zongtao1，2, CHEN Zhe1，2, KANG Jun1，2, ZHU Yishui1，2, TANG Lei1，2, GE Jiandong3, JIANG Hua3. Visualization Analysis Method of Urban Taxi Passenger Travel Characteristics [J]. Computer Engineering and Applications, 2019, 55(6): 237-243.
[6]	DONG Xiwei1，2, WANG Yuwei3, ZHOU Jun1. Robust Multi-View Collaboration Intact Discriminant Subspace Learning Algorithm [J]. Computer Engineering and Applications, 2019, 55(3): 108-114.
[7]	LI Yanqiong, LI Dongdong, WANG Zhe, ZHANG Jing. Multi-View Learning with Gravitational Nearest Neighbor Classifier [J]. Computer Engineering and Applications, 2019, 55(17): 137-142.
[8]	XU Shuang1, YU Li2. Regularized Matrix Factorization Algorithm for Multi-View Data Clustering [J]. Computer Engineering and Applications, 2019, 55(14): 142-147.
[9]	JIA Xiangnan, YU Fengqin, CHEN Ying. Multi-View and Cascaded Regression Model for Face Alignment [J]. Computer Engineering and Applications, 2019, 55(10): 199-204.
[10]	QIAN Yongsheng, SHAO Jie, JI Xinxin, LI Xiaorui, MO Chen, CHENG Qiyu. Multi-view facial expression recognition based on improved convolutional neural network [J]. Computer Engineering and Applications, 2018, 54(24): 12-19.
[11]	YAO Qiong1, XU Xiang1，2, ZOU Kun1. 3D Gabor based multi-view active learning for hyperspectral image classification [J]. Computer Engineering and Applications, 2018, 54(22): 197-204.
[12]	WANG Xianfeng, HUANG Wenzhun, ZHANG Shanwen. Multi-view gait recognition method based on weighted local discriminant canonical correlation analysis [J]. Computer Engineering and Applications, 2018, 54(21): 90-94.
[13]	GUO Weiting, XIA Limin. Human activity recognition based on multi-view nonnegative matrix factorization [J]. Computer Engineering and Applications, 2018, 54(16): 37-43.
[14]	SU Hui1, GE Hongwei1, 2, ZHANG Tao1, YANG Jinlong1 . View correlation factor based multi-view data competition clustering algorithm [J]. Computer Engineering and Applications, 2017, 53(3): 100-105.
[15]	DONG Xiwei. Local manifold reconstruction based semi-supervised multi-view image classification [J]. Computer Engineering and Applications, 2016, 52(18): 24-30.

Improved FGS scalable scheme for multi-view stereo video

改进的多视点立体视频FGS可分级方案

PDF

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 15

Recommended Articles

Metrics