An Efficient Video Similarity Search Algorithm

ChittampallyVasanth Raja
 With the rapid development of modern electronic
equipment, the amount of multimedia data is
increasing tremendously.
 Now a days almost all the digital gadgets are
coming with the in built camera in it.
 Youtube itself contains trillions of videos and
thousands of videos are posted every day all
around the world.
 The rapid increase of multi media video data necessitates an
efficient video similarity search
 There are already many tag based search engines (relying
only on tags not the exact content of video data) ex: Google,
Bing, AltaVista, MSN,Yahoo Search etc.,
 It is a difficult task to retrieve multimedia data
 More computation.. Can We Improve it??
 To solve two challenging problems:
1) similarity measurement
2) search method
 Similarity measurement: The video similarity is measured based
on the calculation of the number of similar video components
 search method: For the scalable computing requirement what
search method do you employ? And What indexing mechanism do you
 Feature extraction: by image characteristic code (ICC) based on the
statistics of spatial temporal distribution.
 Fast Search Approach: for scalable computing was presented
based on clustering index table (CIT)
 Video feature computation is generally based on image
feature extraction.
 Several low-level features such as color, texture, edge are
usually adopted for image fingerprint.
 It has been shown that YCbCr histogram is an effective video
 Advantage: YCbCr coding is widely used in consumer
electronic equipment such as TV, DVR and DVD etc
 The mean of
YCbCr was employed for image feature
 Where M and N are the width and height of image,
respectively. Yij, Cbij,Crij stand for the value of Y, Cb and Cr
components of each pixel
 For video similarity search and noise resistance, the mean
statistics were four digits rounding off integers.
 Image characteristic code (ICC) c is a joint feature
representation made up of three statistical integers of every
pixel components: Y, Cb and Cr. In this way, high dimensional
feature was transformed into compact characteristic code
and video similarity search can be implemented as text
 Image acquisition tool
 Extracted Y, Cb, Cr components from the given image
 Calculated the ICC formula
 Found an interesting point: The average of Y, Cb, Cr
components values of an image are same even when the
image is resized (anti aliasing)
 Extracted frames from the given video
 Can be able to save the frames into hard disk
 Similarity search
 Connecting to the database
 Creating mentioned four tables
[1] An Efficient Video Similarity Search Algorithm. Zheng Cao, Ming
Zhu. IEEE Transactions on Consumer Electronics, Vol. 56, No. 2,
May 2010.
Thank you

similar documents