US 12,073,625 B1
Techniques for generating optimized video segments utilizing a visual search
Najmeh Sadoughi Nourabadi, Bellevue, WA (US); Kewen Chen, Bellevue, WA (US); Tu Anh Ho, Mercer Island, WA (US); Christina Botkins, Seattle, WA (US); Dongqing Zhang, Kirkland, WA (US); and Muhammad Raffay Hamid, Seattle, WA (US)
Assigned to Amazon Technologies, Inc., Seattle, WA (US)
Filed by Amazon Technologies, Inc., Seattle, WA (US)
Filed on Jul. 18, 2023, as Appl. No. 18/223,487.
Application 18/223,487 is a continuation of application No. 17/216,147, filed on Mar. 29, 2021, granted, now 11,763,564.
This patent is subject to a terminal disclaimer.
Int. Cl. G06V 20/40 (2022.01); G06F 16/71 (2019.01); G06F 16/735 (2019.01); G06F 16/75 (2019.01); G06N 20/00 (2019.01)
CPC G06V 20/46 (2022.01) [G06F 16/71 (2019.01); G06F 16/735 (2019.01); G06F 16/75 (2019.01); G06N 20/00 (2019.01); G06V 20/41 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method, comprising:
obtaining, by a computing device, a derivative video segment associated with a portion of video content;
identifying, by the computing device, a starting frame within the video content corresponding to a start of the derivative video segment;
determining, from audio content corresponding to the video content, that the starting frame interrupts dialog occurring at a first time within the audio content corresponding to the starting frame within the video content;
identifying an adjusted starting frame different from the starting frame based at last in part on selecting an earlier video frame occurring prior to the starting frame within the video content, the adjusted starting frame being identified based at least in part on determining that the dialog would be interrupted if the starting frame was selected;
identifying, by the computing device, an ending frame within the video content based at least in part on the adjusted starting frame and a length of the derivative video segment; and
generating, by the computing device, an optimized video segment from the video content based at least in part on: 1) the adjusted starting frame identified from the starting frame of the derivative video segment identified within the video content, and 2) the ending frame identified within the video content.