Autors: Draganov, I. R., Mironov, R. P.
Title: Video Tracing of Moving Objects by Fusing Three-Term Decompositions
Keywords: 3WD, ADMM, Fusion, MAMR, RMAMR, Tensor, TTD, Video tracing

Abstract: In this work, low-rank representation of video with the aim of background modelling and subtraction in order to trace moving objects is investigated based on three-term decompositions. The input video is modelled as a 3-way tensor and over it are applied separately 3-Way-Decomposition (3WD), Motion-Assisted Matrix Restoration (MAMR), Robust Motion-Assisted Matrix Restoration (RMAMR) and the Alternating Direction Method of Multipliers (ADMM). The results from detecting moving objects from the 2 most accurate algorithms (3WD and MAMR) are then combined on a frame basis in order to get more precise results. Two fusing techniques are applied using the logical OR and AND operations. The results are promising and render the proposed algorithms applicable in fields such as video surveillance, vehicle traffic control, crowd monitoring and others.

References

    Issue

    Smart Innovation, Systems and Technologies, vol. 332, pp. 10 - 22, 2023, Singapore, Springer Nature, DOI 10.1007/978-981-19-7842-5_2, ISBN: 978-981197841-8, ISSN: 21903018

    Copyright Springer Nature

    Вид: книга/глава(и) от книга, публикация в издание с импакт фактор, публикация в реферирано издание, индексирана в Scopus