Hyper RPCA: Joint Maximum Correntropy Criterion and Laplacian Scale Mixture Modeling on-the-Fly for Moving Object Detection

Shao, Zerui; Pu, Yi-Fei; Zhou, Jiliu; Wen, Bihan; Zhang, Yi

doi:10.1109/tmm.2021.3121571

Cited by 7 publications

(3 citation statements)

References 67 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, in lq-norm and SSN-based methods, λ is set to an appropriate value for the overall experimental dataset, rather than for each scene. Some experiments 32 – 34 also show that LSM is a better approximation of the l0-norm. Finally, we apply the proposed method to video foreground–background separation.…”

Section: Introductionmentioning

confidence: 93%

Weighted Schatten p-norm and Laplacian scale mixture-based low-rank and sparse decomposition for foreground–background separation

Fan

Jing

et al. 2023

J. Electron. Imag.

View full text Add to dashboard Cite

Low-rank and sparse decomposition (LRSD) plays a vital role in foregroundbackground separation. The existing LRSD methods have the drawback: imprecise surrogate functions of rank and sparsity. We propose the weighted Schatten p-norm (WSN) and Laplacian scale mixture (LSM) method based on LRSD for foreground-background separation, which introduces the WSN and LSM to improve this drawback. To demonstrate the performance of the proposed method, it is applied to foreground-background separation and gets the highest average F-measure score.

show abstract

Section: Introductionmentioning

confidence: 93%

Weighted Schatten p-norm and Laplacian scale mixture-based low-rank and sparse decomposition for foreground–background separation

Fan

Jing

et al. 2023

J. Electron. Imag.

View full text Add to dashboard Cite

show abstract

“…To robustly mitigate the complex noise disturbances, some stable decomposition formulations were developed into three-term decomposition that includes a noise component [46], expressing a single independent identically distributed (i.i.d.) distribution, such as Gaussian and Laplacian [2], [6], or even sparser components [35], while recent works favour mixed noise modelling, such as a mixture of Gaussian (MoG) [6], [37] and an informationtheoretic learning strategy [47], [48], to quantify the noise perturbation. However, these methods cannot model the spatiotemporally distributed signal-dependent noises in the XCA-like heterogeneous environments related to patient and device variability or non-identically distributed data.…”

Section: Rpca-based Foreground/background Separationmentioning

confidence: 99%

“…V IDEO decomposition into foreground/background components is very important for moving object extraction in computer vision, machine learning, and medical imaging [1], [2], [3], [4], [5]. Simply subtracting a static background frame from the current frame may easily lead to incomplete foreground extraction due to the following immediate changes in real scenarios: motion variations of dynamic background [6] and camera, illumination and intensity changes in background/foreground components, and complex noises occurring in low-light images.…”

Section: Introductionmentioning

confidence: 99%

Working memory inspired hierarchical video decomposition with transformative representations

Qin¹,

Mao²,

Zhang³

et al. 2022

Preprint

View full text Add to dashboard Cite

Video decomposition is very important to extract moving foreground objects from complex backgrounds in computer vision, machine learning, and medical imaging, e.g., extracting moving contrast-filled vessels from the complex and noisy backgrounds of X-ray coronary angiography (XCA). However, the challenges caused by dynamic backgrounds, overlapping heterogeneous environments and complex noises still exist in video decomposition. To solve these challenges, this study is the first to introduce a flexible visual working memory model in video decomposition to provide interpretable and high-performance hierarchical deep learning architecture, integrating the transformative representations between sensory and control layers from the perspective of visual and cognitive neuroscience. Specifically, robust PCA unrolling networks acting as a structure-regularized sensor layer decompose XCA into sparse/low-rank structured representations to separate moving contrast-filled vessels from noisy and complex backgrounds. Then, patch recurrent convolutional LSTM networks with a backprojection superresolution module embody unstructured random representations of the control layer in working memory, recurrently projecting spatiotemporally decomposed nonlocal patches into orthogonal subspaces for heterogeneous vessel retrieval and interference suppression. This video decomposition architecture effectively restores the heterogeneous profiles of intensity and geometry of moving objects against the complex background interferences. Experiments show that the proposed method significantly outperforms state-of-the-art methods in accurate moving contrast-filled vessel extraction with excellent flexibility and computational efficiency.

show abstract

Video foreground and background separation via Gaussian scale mixture and generalized nuclear norm based robust principal component analysis

Yang,

2025

Digital Signal Processing

View full text Add to dashboard Cite

Hyper RPCA: Joint Maximum Correntropy Criterion and Laplacian Scale Mixture Modeling on-the-Fly for Moving Object Detection

Cited by 7 publications

References 67 publications

Weighted Schatten p-norm and Laplacian scale mixture-based low-rank and sparse decomposition for foreground–background separation

Weighted Schatten p-norm and Laplacian scale mixture-based low-rank and sparse decomposition for foreground–background separation

Working memory inspired hierarchical video decomposition with transformative representations

Video foreground and background separation via Gaussian scale mixture and generalized nuclear norm based robust principal component analysis

Contact Info

Product

Resources

About