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

Patent metadata

Field	Value
Publication number	US-12322181-B2
Application number	US-202217820359-A
Country	US
Kind code	B2
Filing date	Aug 17, 2022
Priority date	Aug 18, 2021
Publication date	Jun 3, 2025
Grant date	Jun 3, 2025

Abstract

A method for extracting a fingerprint of a video having a plurality of frames includes obtaining a plurality of pixel value matrices from each of the plurality of frames, calculating maximum values of average pixel values in each axis of the plurality of pixel value matrices for each of the plurality of frames, and calculating the fingerprint of the video based on a temporal correlation of the maximum values calculated for the plurality of frames.

First claim

The invention claimed is:

1. A method for extracting a fingerprint of a video having a plurality of frames, the method being executed by one or more processors and comprising:
extracting the plurality of frames from the video;
obtaining a plurality of pixel value matrices from each of the plurality of frames;
calculating maximum values of average pixel values in each axis of the plurality of pixel value matrices for each of the plurality of frames;
calculating the fingerprint of the video based on a temporal correlation of the maximum values calculated for the plurality of frames; and
comparing the fingerprint of the video with the fingerprint of an original video and determining if the fingerprint of the video is within an error range of at least a portion of the fingerprint of the original video.

2. The method according to claim 1 , wherein the calculating of the maximum values of the average pixel values includes:
calculating maximum values of average pixel values in a horizontal axis of the plurality of pixel value matrices for each of the plurality of frames; and
calculating maximum values of average pixel values in a vertical axis of the plurality of pixel value matrices for each of the plurality of frames.

3. The method according to claim 1 , wherein the calculating of the fingerprint of the video includes performing discrete cosine transform (DCT) on the maximum values calculated for the plurality of frames.

4. The method according to claim 3 , wherein the performing of the discrete cosine transform includes:
calculating a first DCT coefficient by performing a discrete cosine transform on the maximum values of the average pixel values in the horizontal axis of the plurality of pixel value matrices in each of the plurality of frames;
calculating a second DCT coefficient by performing a discrete cosine transform on the maximum values of the average pixel values in the vertical axis of the plurality of pixel value matrices in each of the plurality of frames; and
generating the fingerprint of the video by combining the first DCT coefficient and the second DCT coefficient.

5. The method according to claim 3 , wherein the calculating of the fingerprint of the video further includes excluding a coefficient having a basis frequency of 0 from the DCT coefficients calculated by the discrete cosine transform.

6. The method according to claim 1 , wherein the calculating of the maximum value of the average pixel values includes:
calculating the maximum values of the average pixel values in the horizontal axis of the plurality of pixel value matrices, and the maximum values of the average pixel values in the vertical axis of the plurality of pixel value matrices for each of the plurality of frames, and
the calculating of the fingerprint of the video includes:
calculating a first DCT coefficient by performing a discrete cosine transform on the maximum values of the average pixel values in the horizontal axis of the plurality of pixel value matrices in each of the plurality of frames;
calculating a second DCT coefficient by performing a discrete cosine transform on the maximum values of the average pixel values in the vertical axis of the plurality of pixel value matrices in each of the plurality of frames; and
generating the fingerprint of the video by combining the first DCT coefficient and the second DCT coefficient.

7. The method according to claim 6 , wherein the calculating of the fingerprint of the video further includes excluding a coefficient having a basis frequency of 0 from the first DCT coefficient and the second DCT coefficient.

8. The method according to claim 1 , further comprising:
obtaining a fingerprint of an original video; and
determining whether the video is a tampered video by comparing the obtained fingerprint of the original video with the extracted fingerprint of the video.

9. 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 .

10. 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 stored in the memory for extracting a fingerprint of a video, wherein
the one or more programs include instructions for:
extracting the plurality of frames from the video;
obtaining a plurality of pixel value matrices from each of a plurality of frames included in a video;
calculating maximum values of average pixel values in each axis of the plurality of pixel value matrices for each of the plurality of frames;
calculating a fingerprint of the video based on a temporal correlation of the maximum values calculated for the plurality of frames; and
comparing the fingerprint of the video with the fingerprint of an original video and determining if the fingerprint of the video is within an error range of at least a portion of the fingerprint of the original video.

11. The computing device according to claim 10 , wherein the calculating of the maximum values of the average pixel values includes:
calculating maximum values of average pixel values in a horizontal axis of the plurality of pixel value matrices for each of the plurality of frames; and
calculating maximum values of average pixel values in a vertical axis of the plurality of pixel value matrices for each of the plurality of frames.

12. The computing device according to claim 10 , wherein the calculating of the fingerprint of the video includes performing discrete cosine transform on the maximum values calculated for each of the plurality of frames.

13. The computing device according to claim 12 , wherein the performing of the discrete cosine transform includes:
calculating a first DCT coefficient by performing a discrete cosine transform on the maximum values of the average pixel values in the horizontal axis of the plurality of pixel value matrices in each of the plurality of frames;
calculating a second DCT coefficient by performing a discrete cosine transform on the maximum values of the average pixel values in the vertical axis of the plurality of pixel value matrices in each of the plurality of frames; and
generating the fingerprint of the video by combining the first DCT coefficient and the second DCT coefficient.

14. The computing device according to claim 12 , wherein the calculating of the fingerprint of the video includes excluding a coefficient having a basis frequency of 0 from the DCT coefficients calculated by the discrete cosine transform.

15. The computing device according to claim 10 , wherein the calculating of the maximum value of the average pixel values includes:
calculating the maximum values of the average pixel values in the horizontal axis of the plurality of pixel value matrices, and the maximum values of the average pixel values in the vertical axis of the plurality of pixel value matrices for each of the plurality of frames, and
the calculating of the fingerprint of the video includes:
calculating a first DCT coefficient by performing a discrete cosine transform on the maximum values of the average pixel values in the horizontal axis of the plurality of pixel value matrices in each of the plurality of frames;
calculating a second DCT coefficient by performing a discrete cosine transform on the maximum values of the average pixel values in the vertical axis of the plurality of pixel value matrices in each of the plurality of frames; and
generating the fingerprint of the video by combining the first DCT coefficient and the second DCT coefficient.

16

Assignees

• Naver Corp

Inventors

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

Classifications

• G06V20/46 Primary

  Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames · CPC title

• G06V10/753

  Transform-based matching, e.g. Hough transform · CPC title

• G06V20/41

  Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items (segmenting video sequences G06V20/49) · CPC title

• G06T2207/20052

  Discrete cosine transform [DCT] · CPC title

• H04N21/23892

  involving embedding information at multiplex stream level, e.g. embedding a watermark at packet level · CPC title