.png){kind=icon}
.png){kind=icon}
Return
Robust object tracking based on ridge regression and multi-scale local sparse coding
.png){kind=icon}
PRE
AI
DOI：10.1007/s11042-019-08139-2.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Recently, the technology of visual object tracking has achieved great success. However, it is still extraordinary challenging for some factors, such as scale variations, partial occlusions and so on. To deal with the problem of scale variations of the target, this paper proposes a hybrid tracking algorithm based on ridge regression and multi-scale local sparse coding. The hybrid tracking algorithm contains three parts. Firstly, a discriminative model based on two ridge regression models which include a correlation filtering ridge regression model and a color statistics ridge regression model, is used to estimate the approximate position of the target. Secondly, a multi-scale local sparse coding with particle filtering model, which combines local overlapped patches and local non-overlapped patches, is used to estimate the precise position and scale variations of the target. Thirdly, the appearance model of the target in the discriminative model based on ridge regression is updated according to the precise position and scale variations of the target in the second part. At the end, extensive experiments verify the effectiveness of the hybrid tracking algorithm in dealing with scale variations of the target.
Keywords:
Object tracking
Ridge regression
Correlation filtering
Sparse coding
Particle filtering
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
3
Papers:
1.9W
Citations:
3.2W