Summarization of human activity videos via low-rank approximation

Ioannis Mademlis, Anastasios Tefas, Nikos Nikolaidis, Ioannis Pitas

Research output: Chapter in Book/Report/Conference proceedingConference Contribution (Conference Proceeding)

10 Citations (Scopus)
316 Downloads (Pure)

Abstract

Summarization of videos depicting human activities is a timely problem with important applications, e.g., in the domains of surveillance or film/TV production, that steadily becomes more relevant. Research on video summarization has mainly relied on global clustering or local (frame-by-frame) saliency methods to provide automated algorithmic solutions for key-frame extraction. This work presents a method based on selecting as key-frames video frames able to optimally reconstruct the entire video. The novelty lies in modelling the reconstruction algebraically as a Column Subset Selection Problem (CSSP), resulting in extracting key-frames that correspond to elementary visual building blocks. The problem is formulated under an optimization framework
and approximately solved via a genetic algorithm. The proposed video summarization method is being evaluated using a publicly available
annotated dataset and an objective evaluation metric. According to the
quantitative results, it clearly outperforms the typical clustering approach.
Original languageEnglish
Title of host publication2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2017)
Subtitle of host publicationProceedings of a meeting held 5-9 March 2017, New Orleans, Louisiana, USA
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Pages1627-1631
Number of pages5
ISBN (Electronic)9781509041176
ISBN (Print)9781509041183
DOIs
Publication statusPublished - Aug 2017

Publication series

Name
ISSN (Print)2379-190X

Keywords

  • video summarization
  • Sparse dictionary learning
  • Genetic algorithm

Fingerprint Dive into the research topics of 'Summarization of human activity videos via low-rank approximation'. Together they form a unique fingerprint.

Cite this