Method and apparatus for extracting a fingerprint of a video having a plurality of frames

Patent metadata

Field	Value
Publication number	US-12159459-B2
Application number	US-202217820354-A
Country	US
Kind code	B2
Filing date	Aug 17, 2022
Priority date	Aug 18, 2021
Publication date	Dec 3, 2024
Grant date	Dec 3, 2024

Abstract

A method for extracting a fingerprint of a video includes calculating 2D discrete cosine transform (DCT) coefficients from each of the plurality of frames of the video, extracting, from the 2D DCT coefficients, a coefficient having a basis satisfying at least one of up-down symmetry or left-right symmetry, and calculating a fingerprint of the video based on the extracted coefficient.

First claim

The invention claimed is:

1. A method for extracting a fingerprint of a video being identified, the method being executed by one or more processors and comprising:
calculating a 2D discrete cosine transform (DCT) coefficient from each of a plurality of frames of the video;
extracting, from the 2D DCT coefficients, a coefficient having a basis satisfying at least one of up-down symmetry or left-right symmetry; and
calculating a fingerprint of the video based on the extracted coefficient.

2. The method according to claim 1 , wherein the calculating of the 2D DCT coefficient includes:
obtaining a plurality of pixel value matrices from each of the plurality of frames; and
calculating the 2D DCT coefficients from each of the plurality of pixel value matrices.

3. The method according to claim 1 , wherein the calculating of the 2D DCT coefficient includes:
calculating the 2D DCT coefficients from each of the plurality of frames; and
selecting, from the 2D DCT coefficients, a plurality of low-band coefficients having a horizontal frequency of a basis equal to or less than a first predefined frequency, and a vertical frequency of a basis equal to or less than a second predefined frequency.

4. The method according to claim 1 , wherein the extracting of the coefficient having the basis satisfying at least one of up-down symmetry or left-right symmetry includes:
selecting, from the 2D DCT coefficients, a plurality of low-band coefficients having a horizontal frequency of a basis equal to or less than a first predefined frequency, and a vertical frequency of a basis equal to or less than a second predefined frequency; and
extracting, from the plurality of low-band coefficients, a plurality of coefficients having a basis satisfying up-down symmetry or left-right symmetry.

5. The method according to claim 1 , wherein the extracting of the coefficient having the basis satisfying at least one of up-down symmetry or left-right symmetry includes:
selecting, from the 2D DCT coefficients, a plurality of low-band coefficients having a horizontal frequency of a basis equal to or less than a first predefined frequency, and a vertical frequency of a basis equal to or less than a second predefined frequency; and
extracting, from the plurality of low-band coefficients, a plurality of coefficients having a basis satisfying up-down symmetry and left-right symmetry.

6. The method according to claim 1 , wherein the extracting of the coefficient having the basis satisfying at least one of up-down symmetry or left-right symmetry includes excluding, from the 2D DCT coefficients, a coefficient having both a horizontal frequency of a basis and a vertical frequency of a basis of 0.

7. The method according to claim 1 , wherein the calculating of the fingerprint of the video includes calculating, from the extracted coefficients, a mean of variances of the extracted coefficients according to temporal change in the plurality of frames.

8. The method according to claim 7 , wherein the calculating of the mean of variances of the extracted coefficients includes:
calculating a variance between adjacent frames of coefficients having the same basis among the extracted coefficients; and
calculating a mean of the calculated variances for each basis.

9. The method according to claim 1 , further comprising:
obtaining a fingerprint of an original video; and
determining whether or not the video being identified is a tampered video, by comparing the obtained fingerprint of the original video with the extracted fingerprint of the video being identified.

10. A non-transitory computer-readable recording medium storing instructions that, when executed by one or more processors, cause performance of the method according to claim 1 .

11. A computing device, comprising:
a memory; and
one or more processors connected to the memory and configured to execute one or more computer-readable programs included in the memory for extracting a fingerprint of a video being identified, wherein
the one or more programs include instructions for:
calculating 2D DCT coefficients from each of a plurality of frames included in the video;
extracting, from the 2D DCT coefficients, a coefficient having a basis satisfying at least one of up-down symmetry or left-right symmetry; and
calculating a fingerprint of the video based on the extracted coefficient.

12. The computing device according to claim 11 , wherein the calculating of the 2D DCT coefficients includes:
obtaining a plurality of pixel value matrices from each of the plurality of frames; and
calculating the 2D DCT coefficients from each of the plurality of pixel value matrices.

13. The computing device according to claim 11 , wherein the calculating of the 2D DCT coefficients includes:
calculating the 2D DCT coefficients from each of the plurality of frames; and
selecting, from the 2D DCT coefficients, a plurality of low-band coefficients having a horizontal frequency of a basis equal to or less than a first predefined frequency, and a vertical frequency of a basis equal to or less than a second predefined frequency.

14. The computing device according to claim 11 , wherein the extracting of the coefficient having the basis satisfying at least one of up-down symmetry or left-right symmetry includes:
selecting, from the 2D DCT coefficients, a plurality of low-band coefficients having a horizontal frequency of a basis equal to or less than a first predefined frequency, and a vertical frequency of a basis equal to or less than a second predefined frequency; and
extracting, from the plurality of low-band coefficients, a plurality of coefficients having a basis satisfying up-down symmetry or left-right symmetry.

15. The computing device according to claim 11 , wherein the extracting of the coefficient having the basis satisfying at least one of up-down symmetry or left-right symmetry includes:
selecting, from the 2D DCT coefficients, a plurality of low-band coefficients having a horizontal frequency of a basis equal to or less than a first predefined frequency, and a vertical frequency of a basis equal to or less than a second predefined frequency; and
extracting, from the plurality of low-band coefficients, a plurality of coefficients having a basis satisfying up-down symmetry and left-right symmetry.

16. The computing device according to claim 11 , wherein the extracting of the coefficient having the basis satisfying at least one of up-down symmetry or left-right symmetry includes excluding, from the 2D DCT coefficients, a coefficient having both a horizontal frequency of a basis and a vertical frequency of a basis of 0.

17. The computing device according to claim 11 , wherein the calculating of the fingerprint of the video includes calculating, from the extracted coefficients, a mean of variances of the extracted coefficients according to temporal change in the plurality of frames.

18. The computing device according to claim 17 , wherein the calculating of the mean of variances of the extracted coefficients includes:
calculating a variance between adjacent frames of coefficients having the same basis among the extracted coefficients; and
calculating a mean of the calculated variances for each basis.

19. The computing device according to claim 11 , wherein the at least one program further includes instructions for:
obtaining a fingerprint of an original video; and
determining whether or not the video being identified is a tampered video, by comparing the obtained fingerprint of the original video with the extracted f

Assignees

• Naver Corp

Inventors

• Shin Seong Hyeon
• Jeon Jisu
• Kim Dae Hwang
• Park Jongeun
• Kim Seung-Bin

Classifications

• G06T2207/20052

  Discrete cosine transform [DCT] · CPC title

• G06V10/431

  Frequency domain transformation; Autocorrelation · CPC title

• G06V10/62

  relating to a temporal dimension, e.g. time-based feature extraction; Pattern tracking · CPC title

• G06V20/48

  Matching video sequences · CPC title

• G06V20/95

  Pattern authentication; Markers therefor; Forgery detection · CPC title