IP Library Granted Patent US 11,039,205
Granted Patent B2
US 11,039,205 · App. 16/597,698 · Granted Jun 15, 2021

Fake video detection using block chain

Inventors: Xiaoyong Ye (San Mateo, CA); Warren Benedetto (San Mateo, CA)
Assignee: Sony Interactive Entertainment Inc.
H04N21/44008G06N20/00H04L9/0637H04N21/4394H04N21/4662H04N21/8358
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 11,039,205
App. No.
16/597,698
Granted
Jun 15, 2021
Kind
B2
Abstract

Detection of whether a video is a fake video derived from an original video and altered is undertaken using a block chain that either forbids adding to the block chain copies of original videos that have been altered or indicating in the block chain that an altered video has been altered. Image fingerprinting techniques are described for determining whether video sought to be added to block chain has been altered.

Claims (30)

1. An apparatus comprising:

at least one processor configured with instructions executable by the processor to:

access at least one block chain storing video information;

determine whether a first video sought to be added to the block chain is an altered copy of an original video represented in the block chain;

responsive to determining that the first video sought to be added to the block chain is an altered copy of the original video, refuse to add information pertaining to the first video to the block chain, or add the first video to the block chain with an indication that the first video has been altered from the original video;

wherein determining whether the first video sought to be added to the block chain is an altered copy comprises accessing:

at least one frequency transform configured for receiving audio related to a sequence of frames in the first video and configured for outputting a spectrum;

at least one neural network configured for receiving the spectrum and outputting audio feature vectors that represent the audio; and

at least one analysis module trained to learn natural human speech characteristics configured for receiving the audio feature vectors and outputting indication based thereon as to the audio being altered from original audio.

2. The apparatus of claim 1 , wherein the instructions are executable to:

responsive to determining that the first video sought to be added to the block chain is an altered copy of the original video, refuse to add information pertaining to the first video to the block chain.

3. The apparatus of claim 1 , wherein the instructions are executable to:

responsive to determining that the first video sought to be added to the block chain is an altered copy of the original video, add the first video to the block chain with an indication that the first video has been altered from the original video.

4. The apparatus of claim 1 , wherein the instructions are executable to determine whether the first video is an altered copy of the original video at least in part by:

determining whether at least one texture irregularity on a face in the first video or between the face and background in the first video or both is present in the first video; and

determining whether the first video is altered from the original video at least in part based on the texture irregularity.

5. The apparatus of claim 1 , wherein the instructions are executable to determine whether the first video is an altered copy of the original video at least in part by:

using at least one discrete Fourier transform (DFT) for outputting a spectrum representing at least a portion of the first video; and

determining whether the first video is altered from the original video at least in part using the spectrum.

6. The apparatus of claim 5 , wherein the instructions are executable to:

detect at least one irregularity in the spectrum, wherein the irregularity in the spectrum comprises at least one region of brightness that is brighter than a corresponding region in the original video; and

determine that the first video has been altered responsive to detecting the irregularity in the spectrum.

7. The apparatus of claim 1 , wherein the instructions are executable to:

receive a sequence of frames in the first video;

output feature vectors representing whether movement of a person's face as shown in the sequence of frames exhibits natural movement; and

determine whether the sequence of frames is altered from the original video based on the feature vectors.

8. The apparatus of claim 1 , wherein the instructions are executable to:

execute a video hash on the first video;

compare the video hash on the first video with a video hash of the original video represented in the block chain; and

responsive to comparing the video hash on the first video with the video hash of the original video, determine whether the first video is an altered copy of the original video.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Oct 10, 2019
From: YE, XIAOYONG; BENEDETTO, WARREN
To: SONY INTERACTIVE ENTERTAINMENT INC.
Reel/Frame 050685/0322 →
Continuity (1)
Related Publication 20210112306A1 · Apr 15, 2021
Cited By (3)
US 12,526,151 US 12,574,525 US 12,718,063