Lightweight searchable screen video recording

Mattias Marder, Amir Geva, Yaoping Ruan

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

Command logging of maintenance and operation activities of modern computer systems has become an integral component of customer and audit requirements. In recent years, this logging has usually been achieved via desktop video recording. However, the conventional approach of video recording requires high computation overhead, high network bandwidth, and a large storage size. Searching through video files is also a challenge. In this paper, we present a lossy, but text text-preserving, compression scheme that meets these challenges by creating a sparse bitonal image suitable for optical character recognition (OCR). Using our system for auditing, the bitonal image gets stored on a server. Due to the mechanism's text-preserving compression, we can apply OCR off-line to create annotations of each video frame, making the output searchable. Compared to state-of-the-art compression of raw video, our approach can reduce file size by 50-80%, while using CPU and memory resources similar to other methods.

Original languageEnglish
Title of host publication2012 IEEE Visual Communications and Image Processing, VCIP 2012
DOIs
StatePublished - 1 Dec 2012
Externally publishedYes
Event2012 IEEE Visual Communications and Image Processing, VCIP 2012 - San Diego, CA, United States
Duration: 27 Nov 201230 Nov 2012

Publication series

Name2012 IEEE Visual Communications and Image Processing, VCIP 2012

Conference

Conference2012 IEEE Visual Communications and Image Processing, VCIP 2012
Country/TerritoryUnited States
CitySan Diego, CA
Period27/11/1230/11/12

Keywords

  • Text segmentation
  • audit
  • binarization
  • monitoring
  • screen image
  • screen recording

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition

Fingerprint

Dive into the research topics of 'Lightweight searchable screen video recording'. Together they form a unique fingerprint.

Cite this