Video Foreground-Background Separation via Weighted Schatten-p Norm and Structured Sparsity Decomposition

Yufeng Wei; Mingli Jing; Lan Li; Kun Sun; Ruibo Fan

doi:10.3788/LOP202158.0815008

Journals >Laser & Optoelectronics Progress >Volume 58 >Issue 8 >Page 0815008 > Article

Laser & Optoelectronics Progress
Vol. 58, Issue 8, 0815008 (2021)

Video Foreground-Background Separation via Weighted Schatten-p Norm and Structured Sparsity Decomposition

Yufeng Wei¹, Mingli Jing^1、*, Lan Li², Kun Sun¹, and Ruibo Fan¹

Author Affiliations

¹School of Electronic Engineering, Xi'an Shiyou University, Xi'an, Shaanxi 710065, China

²School of Science, Xi'an Shiyou University, Xi'an, Shaanxi 710065, China

show less

DOI: 10.3788/LOP202158.0815008 Cite this Article Set citation alerts

Yufeng Wei, Mingli Jing, Lan Li, Kun Sun, Ruibo Fan. Video Foreground-Background Separation via Weighted Schatten-p Norm and Structured Sparsity Decomposition[J]. Laser & Optoelectronics Progress, 2021, 58(8): 0815008 Copy Citation Text

show less

Structured figures based on 3 pixel×3 pixel overlapping group. (a)(b) Two different foreground distributions; (c)(d) overlapping group of two different distributions

Fig. 1. Structured figures based on 3 pixel×3 pixel overlapping group. (a)(b) Two different foreground distributions; (c)(d) overlapping group of two different distributions

Download full size

Fig. 2. F_measure values under the different C values on Airport dataset

Download full size

Fig. 3. Results of six algorithms on nine sequences

Download full size

Dataset	Video resolution /(pixel×pixel)	Scene feature
Airport	176×144	Dynamic background, changing time information on the frame
Bootstrap	160×120	Static background, some moving people in the room
Curtain	160×128	Dynamic background, the curtains moving slowly and light change gradually
SwitchLight	160×128	Dynamic background, light mutation
WaterSurface	160×128	Dynamic background, water surface fluctuation
Highway	320×240	Dynamic background, the high speed vehicle and the moving shadow
Office	360×240	Static background, a people moving locally with strong light
CamouFlage	160×120	Dynamic background, background with a changing screen
WavingTrees	160×120	Dynamic background, the swinging branch

Table 1. Information of 9 video datasets for comparison

Video	F_measure
Video	GoDec	LSD	PCP	WNNM	WSNM	Proposed
Airport	0.6601	0.7273	0.6601	0.6601	0.6166	0.7103
Bootstrap	0.5110	0.5126	0.5088	0.5088	0.4866	0.5135
Curtain	0.8307	0.8596	0.8309	0.8307	0.8270	0.8614
SwitchLight	0.1268	0.7884	0.7966	0.7989	0.1290	0.1322
WaterSurface	0.8739	0.9170	0.8677	0.8661	0.8412	0.9096
Highway	0.8052	0.8862	0.8887	0.8894	0.8143	0.8869
Office	0.4913	0.4647	0.3777	0.3792	0.4943	0.4648
CamouFlage	0.0000	0.0767	0.1153	0.1146	0.0000	0.9487
WavingTrees	0.7057	0.7213	0.4129	0.4146	0.7081	0.7208
Average	0.5561	0.6615	0.6065	0.6069	0.5463	0.6831

Table 2. F_measure values of six algorithms

Download Citation

Set citation alerts for the article

Tools

Set citation alerts for the article

Save the article for my favorites

Paper Information