Inter-frame video forgery detection based on block-wise brightness variance descriptor

Lu Zheng, Tanfeng Sun, Yun Qing Shi

Research output: Contribution to journalConference articlepeer-review

16 Scopus citations


Video forensics becomes more and more important than ever before. In this paper a new methodology based on Block-wise Brightness Variance Descriptor (BBVD) is proposed. It is capable of fast detecting video inter-frame forgery. Our proposed algorithm has been tested on a database consisting of 240 original and forged videos. The experiments have demonstrated that the precision rate is about 94.09 % in detecting the insertion forgery and the precision rate is 79.45 % in the forgery localization. Moreover, the time utilized for forgery detecting is shorter than the time used for video replay. On average the time of forgery detection is only about 73.4 % in video replay.

Original languageEnglish (US)
Pages (from-to)18-30
Number of pages13
JournalLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
StatePublished - 2015
Event13th International Workshop on Digital-Forensics and Watermarking , IWDW 2014 - Taipei, Taiwan, Province of China
Duration: Oct 1 2014Oct 4 2014

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • General Computer Science


  • Block-wise brightness variance descriptor
  • Inter-frame forgery
  • Video forensics


Dive into the research topics of 'Inter-frame video forgery detection based on block-wise brightness variance descriptor'. Together they form a unique fingerprint.

Cite this