Natural image matting based on surrogate model

Liang, Yihui; Gou, Hongshan; Feng, Fujian; Liu, Guisong; Huang, Han

doi:10.1016/j.asoc.2023.110407

Cited by 5 publications

(5 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When computing resources are limited, it may be challenging to directly use traditional objective function-based methods. Accordingly, we assume that approximating the evaluation of the objective function by fitting the parameters of Gaussian processes (GPs) [ 20 ] can improve the accuracy of alpha matte. Therefore, the GPFM can be represented as follows:

where

is the Gaussian process fitting model (GPFM).…”

Section: Multi-criterion Matting Algorithm Via Gaussian Processmentioning

confidence: 99%

“…Therefore, the GP is completely determined by the mean function and the covariance kernel function between any two random variables [ 37 , 38 ]. In this work, the optimization of the GPFM model is achieved by approximating the optimal solution of the pixel pair objective function, essentially by estimating the undetermined parameters in the GPFM and then using the evaluated parameters to obtain the optimal solution [ 20 ]. Let

and

denote the parameter vector and the estimated parameter vector of the kernel function in the GPFM, respectively.…”

Section: Multi-criterion Matting Algorithm Via Gaussian Processmentioning

confidence: 99%

“…Sampling-based methods [ 18 , 19 ] to evaluate the alpha value of the unknown region by collecting the known foreground and background pixels as candidate samples, which narrows the search range through different sampling strategies. For example, Liang et al [ 20 ] proposed a surrogate model based on natural image matting that effectively reduces computational resource consumption by building on top of random sampling. However, this approach also faces the issue of the potential loss of optimal pixel pairs due to the random sampling strategy.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Multi-Criterion Sampling Matting Algorithm via Gaussian Process

et al. 2023

Self Cite

View full text Add to dashboard Cite

Natural image matting is an essential technique for image processing that enables various applications, such as image synthesis, video editing, and target tracking. However, the existing image matting methods may fail to produce satisfactory results when computing resources are limited. Sampling-based methods can reduce the dimensionality of the decision space and, therefore, reduce computational resources by employing different sampling strategies. While these approaches reduce computational consumption, they may miss an optimal pixel pair when the number of available high-quality pixel pairs is limited. To address this shortcoming, we propose a novel multi-criterion sampling strategy that avoids missing high-quality pixel pairs by incorporating multi-range pixel pair sampling and a high-quality sample selection method. This strategy is employed to develop a multi-criterion matting algorithm via Gaussian process, which searches for the optimal pixel pair by using the Gaussian process fitting model instead of solving the original pixel pair objective function. The experimental results demonstrate that our proposed algorithm outperformed other methods, even with 1% computing resources, and achieved alpha matte results comparable to those yielded by the state-of-the-art optimization algorithms.

show abstract

where

is the Gaussian process fitting model (GPFM).…”

Section: Multi-criterion Matting Algorithm Via Gaussian Processmentioning

confidence: 99%

and

denote the parameter vector and the estimated parameter vector of the kernel function in the GPFM, respectively.…”

Section: Multi-criterion Matting Algorithm Via Gaussian Processmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Multi-Criterion Sampling Matting Algorithm via Gaussian Process

et al. 2023

Self Cite

View full text Add to dashboard Cite

show abstract

“…When computing resources are limited, it may be challenging to directly use traditional objective function-based methods. Accordingly, we assume that approximating the evaluation of the objective function by fitting the parameters of Gaussian processes (GPs) [20] can improve the accuracy of alpha matte. Therefore, the GPFM can be represented as follows:…”

Section: Gaussian Process Fitting Modelmentioning

confidence: 99%

“…Therefore, the GP is completely determined by the mean function and the covariance kernel function between any two random variables [37,38]. In this work, the optimization of the GPFM model is achieved by approximating the optimal solution of the pixel pair objective function, essentially by estimating the undetermined parameters in the GPFM and then using the evaluated parameters to obtain the optimal solution [20]. Let θ and θ denote the parameter vector and the estimated parameter vector of the kernel function in the GPFM, respectively.…”

Section: Gaussian Process Fitting Modelmentioning

confidence: 99%

Multi-criterion Sampling Matting Algorithm via Gaussian Process

Yuan¹,

Gou²,

Tan³

et al. 2023

Preprint

View full text Add to dashboard Cite

Natural image matting is an essential technique for image processing that enables various applications, such as image synthesis, video editing, and target tracking. However, the existing image matting methods may fail to produce satisfactory results when computing resources are limited. Sampling-based methods can reduce the dimensionality of the decision space and therefore reduce computational resources by employing different sampling strategies. While these approaches reduce computational consumption, they may miss an optimal pixel pair when the number of available high-quality pixel pairs is limited. To address this shortcoming, we propose a novel multi-criterion sampling strategy that avoids missing high-quality pixel pairs by incorporating multi-range pixel pair sampling and high-quality samples selection method. This strategy is employed to develop a multi-criterion matting algorithm via Gaussian process, which searches for the optimal pixel pair by using the Gaussian process fitting model instead of solving the original pixel pair objective function. Experimental results demonstrate that our proposed algorithm outperforms other methods even with 1% computing resources, and achieves alpha matte results comparable to those yielded by the state-of-the-art optimization algorithms.

show abstract

Design and optimal scheduling of forecasting-based campus multi-energy complementary energy system

Dong,

Sun,

et al. 2024

Energy

View full text Add to dashboard Cite

Natural image matting based on surrogate model

Cited by 5 publications

References 23 publications

Multi-Criterion Sampling Matting Algorithm via Gaussian Process

Multi-Criterion Sampling Matting Algorithm via Gaussian Process

Multi-criterion Sampling Matting Algorithm via Gaussian Process

Design and optimal scheduling of forecasting-based campus multi-energy complementary energy system

Contact Info

Product

Resources

About