A Simple but Effective Approach to Video Copy Detection

Authors:
Gerhard Roth;Robert Laganière;Patrick Lambert;Ilias Lakhmiri;Tarik Janati
Affiliations:
-;-;-;-;-
Venue:
CRV '10 Proceedings of the 2010 Canadian Conference on Computer and Robot Vision
Year:
2010

Citing 0
Cited 4

Spatio-temporal video copy detection

Proceedings of the 3rd Multimedia Systems Conference
Copy detection of 3D videos

Proceedings of the 3rd Multimedia Systems Conference
Spider: A system for finding 3D video copies

ACM Transactions on Multimedia Computing, Communications, and Applications (TOMCCAP)
A novel spatio-temporal registration framework for video copy localization based on multimodal features

Signal Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Video copy detection is an important task with many applications, especially since detecting copies is an alternative to watermarking. In this paper we describe a simple, but efficient approach that is easy to parallelize, works well, and has low storage requirements. We represent each video frame by a count of the number of SURF interest points in each of 4 by 4 quadrants, a total of 16 bytes per frame. This representation is tolerant of the typical transformations that exist in video, but is still computationally efficient and compact. The approach was tested on the TRECVID copy detection task, for which approximately 15 different groups submitted a solution. Performance was among the best for localization, and was approximately equal to the median with regards to the false positive/negative rate. However, performance varies significantly with the video transformation. We believe that the change in gamma, and decrease in video quality transformations are the most common in practice. For these transformations our method works well.