Detecting Foreground in Videos via Posterior Regularized Robust Bayesian Tensor Factorization

Shenghao Xia, Yinwei Zhang, Biao Zhang, Jian Liu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Foreground detection is a critical step for separating the moving object from the background in video processing. Tensor factorization has been used in foreground detection due to its ability to process complex high-dimensional data, such as color images and videos. However, traditional tensor factorization often lacks the ability for uncertainty quantification. Bayesian tensor factorization can measure the uncertainty by considering the distributions of the tensor factorization model parameters. Besides, domain knowledge is commonly available and could improve the accuracy of foreground detection of the Bayesian tensor factorization model if it can be appropriately incorporated. In this work, a new Bayesian tensor factorization model, named Posterior Regularized Bayesian Robust Tensor Factorization (PR-BRTF), is proposed with incorporating characteristics of dynamic foreground, as a sparsity posterior regularization term. Furthermore, the variational Bayesian inference and L1 norm is combined for inducing sparsity with an efficient inference. The experiments in real-world case studies have shown the performance improvement of the proposed model over state-of-art methods.

Original languageEnglish (US)
Title of host publication2023 IEEE 19th International Conference on Automation Science and Engineering, CASE 2023
PublisherIEEE Computer Society
ISBN (Electronic)9798350320695
DOIs
StatePublished - 2023
Event19th IEEE International Conference on Automation Science and Engineering, CASE 2023 - Auckland, New Zealand
Duration: Aug 26 2023Aug 30 2023

Publication series

NameIEEE International Conference on Automation Science and Engineering
Volume2023-August
ISSN (Print)2161-8070
ISSN (Electronic)2161-8089

Conference

Conference19th IEEE International Conference on Automation Science and Engineering, CASE 2023
Country/TerritoryNew Zealand
CityAuckland
Period8/26/238/30/23

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Detecting Foreground in Videos via Posterior Regularized Robust Bayesian Tensor Factorization'. Together they form a unique fingerprint.

Cite this