US 12,444,214 B1
Authenticity seal for video segments showing a human speaker
Victor Cho, Burlingame, CA (US); Rupali Pathania, Redmond, WA (US); and Samuel Nathan Brown, San Francisco, CA (US)
Assigned to Emovid Corporation, Seattle, WA (US)
Filed by Emovid Corporation, Seattle, WA (US)
Filed on Jun. 5, 2025, as Appl. No. 19/229,872.
Claims priority of provisional application 63/657,470, filed on Jun. 7, 2024.
Int. Cl. G06V 20/00 (2022.01); G06T 11/60 (2006.01)
CPC G06V 20/95 (2022.01) [G06T 11/60 (2013.01)] 18 Claims
OG exemplary drawing
 
13. A method in a first computing system, the method comprising:
transmitting to a second computing system a video sequence data structure, comprising:
a first portion specifying a first audio sequence comprising natural language speech, the first audio sequence corresponding to a second video sequence captured by a microphone;
a second portion specifying a first video sequence synchronized with the first audio sequence of the first portion, the first video sequence corresponding to a second video sequence captured by an image sensor, the second portion comprising a plurality of video frames, each frame comprising:
a first region containing an image of at least the head of a person speaking at a point in the audio sequence corresponding to a position of the frame in the first video sequence, the head comprising hair;
a second region containing a background at least partially surrounding the image of at least the head of a person; and
a third region containing a visual indication, the visual indication comprising:
a first subregion having an appearance that indicates whether the second region has been altered in the first video sequence, relative to the second video sequence;
a second subregion having an appearance that indicates whether the hair has been altered in the first video sequence, relative to the second video sequence; and
a third subregion having an appearance that indicates whether the first audio sequence has been altered, relative to the second audio sequence.