A real-time system for detecting indecent videos based on spatiotemporal patterns

Soonhong Jung, Junsic Youn, Sanghoon Sull

Research output: Contribution to journalArticlepeer-review

4 Citations (Scopus)

Abstract

In this paper, an efficient real-time system is proposed for detecting indecent video scenes, which usually contain periodically moving skin-colored objects. Spatiotemporal motion trajectories are efficiently extracted by the simple color-based region segmentation of a spatiotemporal pattern generated from an input video. By analyzing the vertical displacement of the spatiotemporal motion trajectory along the time axis of the spatiotemporal pattern, the trajectory is then converted to a one-dimensional (1D) signal. Feature vectors computed from the discrete Fourier transform (DFT) of the 1D signal and the colors near the spatiotemporal motion trajectory are used as the inputs for a classifier used to detect adult scenes. Experimental results show improvements in true positive and false alarm rates when compared to existing methods, and significantly reduced processing times.

Original languageEnglish
Article number7027345
Pages (from-to)696-701
Number of pages6
JournalIEEE Transactions on Consumer Electronics
Volume60
Issue number4
DOIs
Publication statusPublished - 2014 Nov 1

Keywords

  • Indecent video
  • periodic motion
  • spatiotemporal motion
  • spatiotemporal pattern

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Media Technology

Fingerprint Dive into the research topics of 'A real-time system for detecting indecent videos based on spatiotemporal patterns'. Together they form a unique fingerprint.

Cite this