US 11,693,902 B2
Relevance-based image selection
Gal Chechik, Los Altos, CA (US); and Samy Bengio, Los Altos, CA (US)
Assigned to GOOGLE LLC, Mountain View, CA (US)
Filed by Google LLC, Mountain View, CA (US)
Filed on May 24, 2021, as Appl. No. 17/328,442.
Application 17/328,442 is a continuation of application No. 16/100,414, filed on Aug. 10, 2018, granted, now 11,017,025, issued on May 25, 2021.
Application 16/100,414 is a continuation of application No. 14/687,116, filed on Apr. 15, 2015, granted, now 10,614,124, issued on Apr. 7, 2020.
Application 14/687,116 is a continuation of application No. 12/546,436, filed on Aug. 24, 2009, abandoned.
Prior Publication US 2021/0349944 A1, Nov. 11, 2021
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 16/78 (2019.01); G06F 16/70 (2019.01); G06F 16/738 (2019.01); G06F 16/74 (2019.01); G06F 16/783 (2019.01)
CPC G06F 16/7867 (2019.01) [G06F 16/70 (2019.01); G06F 16/738 (2019.01); G06F 16/743 (2019.01); G06F 16/78 (2019.01); G06F 16/783 (2019.01); G06F 16/7844 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method for providing searchable videos, the method executed by a computer system comprising one or more processors, and comprising:
storing a searchable video index associated with a plurality of videos that each include a plurality of frames, the searchable video index including a mapping between two or more frames of each of the plurality of videos and one or more keyword representations, the mapping generated based at least in part on a machine-learned model trained to learn correlations between visual content of individual video frames and keyword representations;
receiving a search query including one or more textual keywords;
selecting a first video from the searchable video index based on the one or more textual keywords;
determining a frame of the first video based at least in part on the one or more textual keywords and the mapping between the two or more frames of each of the plurality of videos and the one or more keyword representations; and
providing a result set including the frame of the first video and a link to the first video in response to the search query.