Browse Prior Art Database

A method to compensate significance of the indexes using context of the video and additional information about the indexes

IP.com Disclosure Number: IPCOM000012688D
Original Publication Date: 2003-May-21
Included in the Prior Art Database: 2003-May-21
Document File: 2 page(s) / 85K

Publishing Venue

IBM

Abstract

Disclosed is a method for using predefined some state variables and history of their state transition and dynamically calculating significance of each index at a time point with the function which is given through learning many high-quality video digests generated by professional video editors.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 56% of the total text.

Page 1 of 2

  A method to compensate significance of the indexes using context of the video and additional information about the indexes

  A method to compensate significance of the indexes using context of the video and additional information about the indexes

Disclosed is a method for using predefined some state variables and history of their state transition and dynamically calculating significance of each index at a time point with the function which is given through learning many high-quality video digests generated by professional video editors.

This method requires that all indexes and their additional information for the source video are generated by preprocessing. Indexes are subdivided into following three categories in advance.

Parameters for state decision ( )

Parameters with state-independent significance ( )

Parameters with state-dependent significance ( )

States ( ) which consist of combinations of parameters for state decision and possible

transitions between two states ( ) is also predefined. Significance of the index on the

transition ( ) are calculated by the following function:

--- (i)

where is predefined basic significance of , is the significance function of ,

is the significance function of which is dependent on the

transition and the history of states and is given through learning many high-quality video digests generated by professional video editors.

Learning procedure consists of the following four steps (See Fig. 1).

Time slots in source video are identified by analyzing video digests. Errors within

the allowable limits are accepted. The states in each time slot are determin...