Method of indexing and searching moving picture data using a motion activity description method

Claim 2

The method of claim 1, wherein said video name has a plurality of time stamps and a f_code.

Claim 3

A moving picture indexing and searching method using a motion activity describing method, comprising the steps of: describing a motion feature information in terms of a video name and a time stamp, and a frequency of an Inter mode of No_MC, a frequency of an Intra mode of No_MC and a frequency of an MC mode obtained by a selected process; and indexing and searching a moving picture using said motion feature information.

Claim 4

The method of claim 3, wherein said video name has a plurality of time stamps, a frequency of an Inter mode of a No_MC, a frequency of an Intra mode of a No_MC, and a frequency of an MC mode.

Claim 5

The method of claim 3, wherein said selected process comprises: a first step of determining whether an MB_Mode is a No_MC mode; a second step of determining whether MB_Mode is No_MC_inter mode, if MB_Mode is determined as No_MC mode in said first step; a third step of increasing counting value(no_mc_inter) of a counter for counting a frequency of No_MC_Inter mode and determining whether a current MB is a last MB, if MB_Mode is determined as No_MC_Inter mode in said second step, and determining whether MB_Mode is No_MC_Intra mode if MB_Mode is not determined as No_MC_Inter mode; a fourth step of increasing counting value(no_mc_intra) of a counter for counting a frequency of No_MC_Intra mode and determining whether a current MB is a last MB, if MB_Mode is determined as No_MC_Intra mode in said third step, and increasing counting value(mc) of a counter for counting a frequency of MC mode and determining whether a current MB is a last MB, if MB_Mode is not determined as No_MC mode in said third step; and a fifth step of increasing number of macro blocks (TotMBs) used in a motion feature range and repeatedly performing said first to third steps if said current MB is not determined as a last MB in said fourth step, and obtaining a frequency(No_MC_Inter_Ratio) of Inter mode of No_MC, frequency (Intra_Ratio) of Intra mode of No_MC, and frequency (Intra_Ratio) of Intra mode of No MC, if said current MB is determined as a last MB in said fourth step. [Equation] No_MC_Inter_Ratio = no_mc_inter/TotMBs Intra_Ratio = no_mc_intra/TotMBs MC_Ratio = mc/TotMBs

Claim 6

A moving picture indexing and searching method using a motion activity describing technique, comprising the steps of: describing a motion feature information based on a video name, a time stamp and a frequency of an Inter mode of a No_MC obtained by a selected process; and indexing and searching a moving picture using said motion feature information.

Claim 7

The method of claim 6, wherein said video name has a plurality of time stamps and a frequency of an Inter mode of an No_MC.

Claim 8

The method of claim 6, wherein said selected process comprises: a first step of increasing counting value(mc_nc_inter) of a counter for counting a frequency of No_MC_Inter mode and determining whether a current MB is a last MB if said MB_Mode is No_MC mode and No_MC_Inter mode; and a second step of increasing number of macroblocks (TotMBs) used in a motion feature range and repeatedly performing said first step if said current MB is not determined as a last MB in said first step, and obtaining a frequency(No_MC_Inter_Ratio) of Inter mode of No_MC according to the following Equation, if said current MB is the last MB. [Equation] No_MC_Inter_Ratio = no_mc_inter/TotMBs

Claim 9

A moving picture indexing and searching method using a motion activity describing technique, comprising the steps of: describing a motion feature information based on a video name, a time stamp and a frequency of an Intra mode of a No_MC obtained by a selected process; and indexing and searching a moving picture using said motion feature information.

Claim 10

The method of claim 9, wherein said video name has a plurality of name information and a frequency of Intra mode of a No_MC.

Claim 11

The method of claim 9, wherein said selected process comprises: a first step of increasing counting value(mc_nc_intra) of a counter for counting a frequency of No_MC_Intra mode and determining whether a current MB is a last MB if said MB_Mode is No_MC mode and No_MC_Intra mode; and a second step of increasing number of macro blocks (TotMBs) used in a motion feature range and repeatedly performing said first step if said current MB is not determined as a last MB in said first step, and obtaining frequency(Intra_Ratio) of Intra mode of No_MC according to the following Equation, if said current MB is the last MB. [Equation] Intra_Ratio = no_mc_intra/TotMBs

Claim 12

A moving picture indexing and searching method using a motion activity describing technique, comprising the steps of: describing a motion feature information based on a video name, a time stamp and a frequency of an MC mode obtained by a selected process; and indexing and searching a moving picture using said motion feature information.

Claim 13

The method of claim 12, wherein said video name has a frequency of a plurality of time stamps and a frequency of an MC mode.

Claim 14

The method of claim 12, wherein said selected process comprises: a first step of increasing counting value(mc) of a counter for counting a frequency of MC mode and determining whether a current MB is a last MB if said MB_Mode is not No_MC mode; and a second step of increasing number of macro blocks (TotMBs) used in a motion feature range and repeatedly performing said first step if said current MB is not determined as a last MB in said first step, and obtaining a frequency(MC_Ratio) with respect to MC mode according to the following Equation, if said current MB is the last MB. [Equation] MC_Ratio = mc/TotMBs

Claim 15

A moving picture indexing and searching method using a motion activity describing technique, comprising the steps of: describing a feature information based on a video name and a time stamp and a frequency of an MC_Coded mode and a frequency of an MC_Not_coded mode obtained by a selected process; and indexing and searching a moving picture using said feature information.

Claim 16

The method of claim 15, wherein said video name has a plurality of time stamps, a frequency of an MC_coded mode and a frequency of an MC_Not_Coded mode.

Claim 17

The method of claim 15, wherein said selected process comprises: a first step of determining whether MB_Mode is MC mode; a second step of determining whether MB_Mode is MC_Coded mode, if said MB_Mode is determined as MC mode in said first step; a third step of determining whether a current MB is a last MB by increasing counting value (mc_not_coded) of a counter for counting a frequency of MC_Not_Coded mode, if said MB_Mode is not determined as MC_Coded mode in said second step, and increasing counting value(mc_coded) of a counter for counting a frequency of MC_Coded mode and further increasing counting value(mc) of a counter for counting a frequency of MC mode; a fourth step of increasing number of macro blocks (TotMBs) used in a motion feature range and repeatedly performing said first to third steps if said current MB is determined as not a last MB in said third step, and obtaining a frequency(MC_Coded_Ratio) of the MC_Coded mode and a frequency(MC_Not_Coded_ratio) of the Not_Coded mode according to the following Equation, if said current MB is the last MB. [Equation] MC_Coded_Ratio = mc_coded/mc MC_Not_Coded_Ratio = mc_not_coded/mc

Claim 18

A moving picture indexing and searching method using a motion activity technique, comprising the steps of: describing a feature information based on a video name, a time stamp and a frequency of an MC_Coded mode and a frequency of an MC_Not_Coded mode; and measuring a similarity by comparing the frequency of MC_Coded mode of a moving picture of a smaller f_code with the frequency of MC_Coded mode of a larger f_code or with the frequency of MC_Not_Coded mode with respect to a moving picture having different f_codes, and indexing and searching a moving picture using said feature information.

Claim 19

The method of claim 18, wherein said video name has a plurality of time stamps, a frequency of MC_Coded mode and a frequency of MC_Not_Coded mode.

Claim 20

The method of claim 18, wherein said selected process comprises: a first step of determining whether MB_Mode is MC mode; a second step of determining whether said MB_Mode is MC_Coded mode, if said MB_Mode is determined as MC mode in said first step; a third step of increasing counting value(mc_not_coded) of a counter for counting a frequency of MC_Not_Coded mode if said MB_Mode is not determined as MC_Coded mode in said second step, and increasing counting value(mc_coded) of a counter for counting a frequency of MC_Coded mode, increasing counting value(mc) of a counter for counting MC mode, and determining whether a current MB is a last MB if said MB_Mode is determined as MC_Coded mode in said second step; and a fourth step of increasing number of macro blocks (TotMBs) used in a motion feature range and repeatedly performing said first to third steps if said current MB is not determined as a last MB in said third step, and obtaining a frequency (MC_Coded_Ratio) of MC_Coded mode and a frequency(MC_Not_Coded_Ratio)of Not_Coded mode according to the following Equation if said MB is determined as a last MB in said third step. [Equation] MC_Coded_Ratio = mc_coded/mc MC_Not_Coded_Ratio = mc_not_coded/mc

Claim 21

A moving picture indexing and searching method using a motion activity describing technique, comprising the steps of: describing a motion feature information in terms of a video name, a time stamp and a frequency of Inter mode of No_MC obtained by a selected process, a frequency of Intra mode of said No_MC, a frequency of MC mode, a frequency of a Coded mode with respect to MC mode and a frequency of Not_Coded mode with respect to MC mode; and indexing and searching a moving picture using said motion feature information.

Claim 22

The method of claim 21, wherein said video name has a frequency of a plurality of time stamps, a frequency of Inter mode of No_MC, a frequency of Intra mode of No_MC, a frequency of MC mode, a frequency of Coded mode with respect to Inter mode, a frequency of MC mode, a frequency of Coded mode with respect to MC mode, and a frequency of Not_Coded mode with respect to MC mode.

Claim 23

The method of claim 21, wherein said selected process comprises: a first step of determining whether MB_Mode is No_MC mode; a second step of determining whether said MB_Mode is No_MC inter mode, if said MB_Mode is determined as No_MC mode in said first step; a third step of increasing counting value (no_mc_inter) of a counter for counting a frequency of No_MC_Inter mode if said MB_Mode is determined as No_MC_Inter mode in said second step, and determining whether MB_Mode is No_MC_Intra mode if said MB_Mode is not determined as No_MC_Inter mode in said second step; a fourth step of increasing counting value(no_mc_intra) of a counter for counting a frequency of No_MC_Intra mode if said MB_Mode is determined as No_MC_Intra mode in said third step, and determining whether MB_Mode is MC mode if said MB_Mode is not determined as No_MC_Intra mode; a fifth step of determining whether said MB_Mode is MC_Coded mode if said MB_Mode is determined as MC mode in said fourth step; a sixth step of increasing counting value(mc_not_coded) of a counter for counting a frequency of MC_Not_Coded mode if said MB_Mode is not determined as MC_Coded mode in said fifth step, and increasing counting value(mc_coded) of a counter for counting a frequency of MC_Coded mode, increasing counting value(mc) of a counter for counting a frequency of MC mode, and determining whether a current MB is a 1st MB if said MB_Mode is determined as MC_Coded mode in said fifth step;and a seventh step of increasing number of macro blocks (TotMBs) used in a motion feature range and repeatedly performing said first to sixth steps if said current MB is not determined as a last MB in said sixth step, and obtaining a frequency(No_MC_Inter_Ratio) of Inter mode of No_MC, a frequency(Intra_Ratio) of Intra mode of No_MC, a frequency (MC_Ratio) with respect to MC mode, a frequency(MC_Coded_Ratio) of MC_Coded mode, and a frequency(MC_Not_Coded_Ratio) of Not_Coded mode according to the following Equation if said MB is determined as a last MB in said sixth step. [Equation] No_MC_Inter_Ratio = no_mc_inter/TotMBs Intra_Ratio = no_mc_intra/totMBs MC_Ratio = mc/TotMBs MC_Coded_Ratio = mc_coded/mc MC_Not_Coded_Ratio = mc_not_coded/mc

Claim 24

A moving picture indexing and searching method using a motion activity describing technique, comprising the steps of: classifying a motion vector of a macro block having a video name, time stamp, MC_Not_Coded mode, MC_Coded mode and No_MC_Inter mode in accordance with f_code, describing in terms of f_code frequency obtained by a selected process, measuring a similarity by comparing each f_code frequency, and indexing and searching a moving picture using said motion feature information.

Claim 25

The method of claim 24, wherein said selected step comprises: a first step of determining whether the MB_Mode is an MC_Mode; a second step of increasing a counting value (f_code_c[1]) of a counter of f_code 1 and determining whether a current MB is a last MB if the MB_Mode is determined as Inter mode of No_MC mode; a third step of selecting the largest X-axis motion vector (mv_x) and the largest Y-axis motion vector (mv_y) as a largest motion vector (max_mv), obtaining a maximum value k of f_code by the selected max_mv/16, increasing the counting value (f_code_c[k]) of the obtained f_counter, and determining whether a current MB is the last MB if MB_Mode is determined as MC_Mode in said first step; and a fourth step of increasing a number of macro block used in a motion feature range, repeatedly performing said first to third steps, and obtaining a frequency of f_code according to the following Equation if a current MB is the last MB [Equation] f_code_ratio[k] = f_code_c[k]TotMBs

Claim 26

A moving picture indexing and searching method using a motion activity describing technique, comprising the steps of: classifying a motion vector of a macro block having a video name, time stamp, MC_Not_Coded mode, MC_Coded mode and No_MC_Inter mode in accordance with f_code, describing in terms of f_code frequency obtained by a selected process, measuring a similarity by comparing each f_code frequency with an at least one adjacent f_code frequency, and indexing and searching a moving picture using said motion feature information.

Claim 27

The method of claim 26, wherein said selected step comprises: a first step of determining whether the MB_Mode is an MC_Mode; a second step of increasing a counting value (f_code_c[1]) of a counter of f_code 1 and determining whether a current MB is a last MB if the MB_Mode is determined as Inter mode of No_MC mode; a third step of selecting the largest X-axis motion vector (mv_x) and the largest Y-axis motion vector(mv_y) as a largest motion vector(max_mv), obtaining a maximum value k of f_code by the selected max_mv/16, increasing the counting value (f_code_c[k]) of the obtained f_counter, and determining whether a current MB is the last MB if MB_Mode is determined as MC_Mode in said first step; and a fourth step of increasing a number of macro block used in a motion feature range, repeatedly performing said first to third steps, and obtaining a frequency of f_code according to the following Equation if a current MB is not determined as a last MB in said second and third steps. [Equation] f_code_ratio[k] = f_code_c[k]/TotMBs

Claim 28

A moving picture indexing and searching method using a motion activity describing technique, comprising the steps of: classifying a motion vector of a macro block having a video name, time stamp, MC_Not_Coded mode, MC_Coded mode and No_MC_Inter mode in accordance with f_code, describing in terms of f_code frequency obtained by a selected process, measuring a similarity by comparing each f_code frequency with an at least one adjacent f_code frequency, said comparison is performed in such a manner that said similarity is larger when said motion vector is closer to a maximum value and a minimum value of a vector range described by each f_code and a reproduced motion compensation error is larger, and a weighted value is added to a frequency of f_code larger than the releant f_code, and indexing and searching a moving picture using said f_code frequency.

Claim 29

The method of claim 28, wherein said selected step comprises: a first step of determining whether the MB_Mode is an MC_Mode; a second step of increasing a counting value(f_code_c[1]) of a counter of f_code 1 and determining whether a current MB is a last MB if the MB_Mode is determined as Inter mode of No_MC mode in said first step; a third step of selecting the largest X-axis motion vector (mv_x) and the largest Y-axis motion vector(mv_y) as a largest motion vector(max_mv), obtaining a maximum value k of f_code by the selected max_mv/16, increasing the counting value (f_code_c[k]) of the obtained f_counter, and determining whether a current MB is the last MB if MB_Mode is determined as MC_Mode in said first step; and a fourth step of increasing a number of macro blocks used in a motion feature range and repeatedly performing said first to third steps, and obtaining a frequency of f_code according to the following Equation if the current MB is determined as a last MB in said second and third steps. [Equation] f_code_ratio[k] = f_code_c[k]/TotMBs

Claim 30

A moving picture indexing and searching method using a motion activity describing technique, comprising the steps of: describing a feature information in terms of a video name and a time stamp and a frequency of Inter mode of No_MC and a frequency of Intra mode of No_MC obtained by a selected process; and indexing and searching a moving picture using said feature information.

Claim 31

The method of claim 30, wherein said selected process comprises: a first step of determining whether MC_Mode is No_MC mode; a second step of determining whether said MB_Mode is No_MC_inter mode, if said MC_Mode is determined as No_MC mode in said first step; a third step of increasing counting value(no_mc_inter) of a counter for counting a frequency of No_MC_Inter mode if said MB_Mode is determined as No_MC_Inter mode and determining whether a current MB is a last MB in said second step, and determining whether MB_Mode is No_MC_Intra mode if said MB_Mode is not determined as No_MC_Inter mode in said second step; a fourth step of increasing counting value(no_mc_intra) of a counter for counting a frequency of No_MC_Intra mode if said MB_Mode is determined as No_MC_Intra mode in said third step, and determining whether a current MB is a last MB if said MB_Mode is not determined as No_MC_Intra mode; and a fifth step of increasing number of macro blocks (TotMBs) used in a motion feature range and repeatedly performing said first to fourth steps if said current MB is not determined as a last MB in said third and fourth steps, and obtaining a frequency(No_MC_Inter_Ratio) of Inter mode of No_MC and a frequency(No_MC_Intra_Ratio) of Intra mode of No_MC according to the following Equation if said MB is determined as a last MB in said fourth step. [Equation] No_MC_Inter_Ratio = no_mc_inter/TotMBs Intra_Ratio = no_mc_intra/TotMBs