Identifying altered digital video content

• US 10,631,036 B1
• Filed: 10/24/2018
• Issued: 04/21/2020
• Est. Priority Date: 10/24/2018
• Status: Active Grant

• Current Assignee: Snap, Inc.
• Original Assignee: GfyCat, Inc. (Snap, Inc.)
• Inventors: Rabbat, Richard, Fu, Ernestine
• Primary Examiner(s): Caschera, Antonio A
• Application Number:
  US16/169,804
• Publication Number:
  US 20200137447A1
• Time in Patent Office: 545 Days
• CPC Class Codes: G06F 18/22 Matching criteria, e.g. proximity measures
• CPC Class Codes: G06F 21/51 at application loading time, e.g. accepting, rejecting, starting or inhibiting executable software based on integrity or source reliability
• CPC Class Codes: G06F 21/64 Protecting data integrity, e.g. using checksums, certificates or signatures
• CPC Class Codes: G06V 10/761 Proximity, similarity or dissimilarity measures
• CPC Class Codes: G06V 20/48 Matching video sequences
• CPC Class Codes: G06V 40/161 Detection; Localisation; Normalisation
• CPC Class Codes: G06V 40/172 Classification, e.g. identification
• CPC Class Codes: H04N 21/23418 involving operations for analysing video streams, e.g. detecting features or characteristics television picture signal circuitry for scene change detection H04N5/147; filtering for image enhancement G06T5/00; methods or arrangements for recognising scenes G06V20/00; arrangements characterised by components specially adapted for monitoring, identification or recognition of video in broadcast systems H04H60/59
• CPC Class Codes: H04N 21/4318 by altering the content in the rendering process, e.g. blanking, blurring or masking an image region image enhancement or restoration in general G06T5/00
• CPC Class Codes: H04N 21/44008 involving operations for analysing video streams, e.g. detecting features or characteristics in the video stream arrangements characterised by components specially adapted for monitoring, identification or recognition of video in broadcast systems H04H60/59
• CPC Class Codes: H04N 21/440245 the reformatting operation being performed only on part of the stream, e.g. a region of the image or a time segment
• CPC Class Codes: H04N 21/45455 applied to a region of the image
• CPC Class Codes: H04N 21/835 Generation of protective data, e.g. certificates protecting software against unauthorised usage in a vending or licensing environment G06F21/10

Abstract

In a computer-implemented method for identifying altered digital video content, a digital video content item is accessed. A visually identifiable region is identified within the digital video content item. A mask is applied to the visually identifiable region of the digital video content item, wherein the mask blocks the visually identifiable region of the digital video content item. The digital video content item with the mask is compared to other digital video content items, wherein the comparing disregards the visually identifiable region according to the mask. Provided the digital video content is identified as similar to at least one other digital video content item, the visually identifiable region of the digital content video item is compared to a visually identifiable region of the at least one other digital video content item. Provided the visually identifiable region of the digital content video item does not match the visually identifiable region of the at least one other digital video content item, it is determined that the digital video content item is altered.

Claims

1. A computer-implemented method for identifying altered digital video content, the method comprising:

accessing, at a computer system, a digital video content item;

identifying, at the computer system, a visually identifiable region within the digital video content item;

applying, at the computer system, a mask to the visually identifiable region of the digital video content item, wherein the mask blocks the visually identifiable region of the digital video content item;

comparing, at the computer system, the digital video content item with the mask to other digital video content items, wherein the comparing disregards the visually identifiable region according to the mask;

provided the digital video content is identified as similar to at least one other digital video content item, comparing, at the computer system, the visually identifiable region of the digital content video item to a visually identifiable region of the at least one other digital video content item; and

provided the visually identifiable region of the digital content video item does not match the visually identifiable region of the at least one other digital video content item, determining, at the computer system, that the digital video content item is altered.

2. The method of claim 1, wherein the comparing, at the computer system, the digital video content item with the mask to other digital video content items comprises:

generating a plurality of perceptual hashes for frames of the digital video content item; and

comparing the plurality of perceptual hashes for frames of the digital video content item to perceptual hashes for frames of the other digital video content items.

3. The method of claim 2, wherein the comparing, at the computer system, the digital video content item with the mask to other digital video content items further comprises:

determining distances between the perceptual hashes for frames of the digital video content item and the perceptual hashes for frames of the other digital video content items; and

provided at least a portion of the distances satisfy a distance threshold between the digital video content item and one other digital video content item, identifying the digital video content item as similar to the one other digital video content item.

4. The method of claim 1, wherein the comparing, at the computer system, the visually identifiable region of the digital content video item to the visually identifiable region of the at least one other digital video content item comprises:

generating a plurality of perceptual hashes corresponding to the visually identifiable region for frames of the digital video content item; and

comparing the plurality of perceptual hashes corresponding to the visually identifiable region for frames of the digital video content item to perceptual hashes corresponding to the visually identifiable region for frames of the at least one other digital video content items.

5. The method of claim 4, wherein the comparing, at the computer system, the visually identifiable region of the digital content video item to the visually identifiable region of the at least one other digital video content item further comprises:

generating the plurality of perceptual hashes for the visually identifiable region for frames of the at least one other digital video content item.

6. The method of claim 1, further comprising:

responsive to determining that the digital video content item is altered, forwarding, at the computer system, the digital video content item to a human operator for visual confirmation that the digital video content item is altered.

7. The method of claim 6, further comprising:

responsive to receiving a confirmation from the human operator that the digital video content item is altered, using, at the computer system, the digital video content item for training automated alteration analysis of digital video content.

8. The method of claim 1, wherein the visually identifiable region comprises a face.

9. The method of claim 8, wherein the identifying, at the computer system, the visually identifiable region within the digital video content item comprises:

using facial recognition to automatically identify the visually identifiable region comprising the face.

10. The method of claim 1, wherein the digital video content item is a short form looping prerecorded video file.

11. A non-transitory computer readable storage medium having computer readable program code stored thereon for causing a computer system to perform a method for identifying altered digital video content, the method comprising:

accessing a digital video content item;

automatically identifying a facial region within the digital video content item using facial recognition;

applying a mask to the facial region of the digital video content item, wherein the mask blocks the facial region of the digital video content item;

comparing the digital video content item with the mask to other digital video content items, wherein the comparing disregards the facial region according to the mask;

provided the digital video content is identified as similar to at least one other digital video content item, comparing the facial region of the digital content video item to a facial region of the at least one other digital video content item; and

provided the facial region of the digital content video item does not match the facial region of the at least one other digital video content item, determining that the digital video content item is altered.

12. The non-transitory computer readable storage medium of claim 11, wherein the comparing the digital video content item with the mask to other digital video content items comprises:

generating a plurality of perceptual hashes for frames of the digital video content item; and

comparing the plurality of perceptual hashes for frames of the digital video content item to perceptual hashes for frames of the other digital video content items.

13. The non-transitory computer readable storage medium of claim 12, wherein the comparing the digital video content item with the mask to other digital video content items further comprises:

determining distances between the perceptual hashes for frames of the digital video content item and the perceptual hashes for frames of the other digital video content items; and

provided at least a portion of the distances satisfy a distance threshold between the digital video content item and one other digital video content item, identifying the digital video content item as similar to the one other digital video content item.

14. The non-transitory computer readable storage medium of claim 11, wherein the comparing the facial region of the digital content video item to the facial region of the at least one other digital video content item comprises:

generating a plurality of perceptual hashes corresponding to the facial region for frames of the digital video content item; and

comparing the plurality of perceptual hashes corresponding to the facial region for frames of the digital video content item to perceptual hashes corresponding to the facial region for frames of the at least one other digital video content items.

15. The non-transitory computer readable storage medium of claim 14, wherein the comparing the facial region of the digital content video item to the facial region of the at least one other digital video content item further comprises:

generating the plurality of perceptual hashes for the facial region for frames of the at least one other digital video content item.

16. The non-transitory computer readable storage medium of claim 11, the method further comprising:

responsive to determining that the digital video content item is altered, forwarding the digital video content item to a human operator for visual confirmation that the digital video content item is altered.

17. The non-transitory computer readable storage medium of claim 16, the method further comprising:

responsive to receiving a confirmation from the human operator that the digital video content item is altered, using the digital video content item for training automated alteration analysis of digital video content.

18. The non-transitory computer readable storage medium of claim 11, wherein the digital video content item is a short form looping prerecorded video file.

19. A computer system comprising:

a data storage unit; and

a processor coupled with the data storage unit, the processor configured to;

access a digital video content item;

identify a visually identifiable region within the digital video content item;

apply a mask to the visually identifiable region of the digital video content item, wherein the mask blocks the visually identifiable region of the digital video content item;

compare the digital video content item with the mask to other digital video content items while disregarding the visually identifiable region according to the mask;

provided the digital video content is identified as similar to at least one other digital video content item, compare the visually identifiable region of the digital content video item to a visually identifiable region of the at least one other digital video content item; and

provided the visually identifiable region of the digital content video item does not match the visually identifiable region of the at least one other digital video content item, determine that the digital video content item is altered.

20. The computer system of claim 19, wherein the processor is further configured to:

generate a plurality of perceptual hashes for frames of the digital video content item;

compare the plurality of perceptual hashes for frames of the digital video content item to perceptual hashes for frames of the other digital video content items;

determine distances between the perceptual hashes for frames of the digital video content item and the perceptual hashes for frames of the other digital video content items; and

provided at least a portion of the distances satisfy a distance threshold between the digital video content item and one other digital video content item, identify the digital video content item as similar to the one other digital video content item.