US 12,014,542 B2
Selecting and presenting representative frames for video previews
Sanketh Shetty, Sunnyvale, CA (US); Tomas Izo, San Francisco, CA (US); Min-Hsuan Tsai, Grapevine, TX (US); Sudheendra Vijayanarasimhan, Mountain View, CA (US); Apostol Natsev, Sunnyvale, CA (US); Sami Abu-El-Haija, San Francisco, CA (US); George Dan Toderici, Mountain View, CA (US); Susana Ricco, Redwood City, CA (US); Balakrishnan Varadarajan, Sunnyvale, CA (US); Nicola Muscettola, Sunnyvale, CA (US); WeiHsin Gu, Redwood City, CA (US); Weilong Yang, Fremont, CA (US); Nitin Khandelwal, Milpitas, CA (US); and Phuong Le, Sunnyvale, CA (US)
Assigned to Google LLC, Mountain View, CA (US)
Filed by Google LLC, Mountain View, CA (US)
Filed on Dec. 14, 2020, as Appl. No. 17/120,525.
Application 17/120,525 is a continuation of application No. 15/959,858, filed on Apr. 23, 2018, granted, now 10,867,183.
Application 15/959,858 is a continuation of application No. 14/848,216, filed on Sep. 8, 2015, granted, now 9,953,222, issued on Apr. 24, 2018.
Claims priority of provisional application 62/120,107, filed on Feb. 24, 2015.
Claims priority of provisional application 62/047,639, filed on Sep. 8, 2014.
Prior Publication US 2021/0166035 A1, Jun. 3, 2021
This patent is subject to a terminal disclaimer.
Int. Cl. G06K 9/00 (2022.01); G06F 16/783 (2019.01); G06V 20/40 (2022.01)
CPC G06V 20/41 (2022.01) [G06F 16/7834 (2019.01); G06V 20/46 (2022.01); G06V 20/47 (2022.01); G06V 20/49 (2022.01); G06V 2201/10 (2022.01)] 21 Claims
OG exemplary drawing
 
1. A computer-implemented method for selecting representative frames for videos, the method comprising:
receiving a search query from a user of a user device;
identifying a plurality of semantic features for one or more frames of a video by determining, using a plurality of semantic classifiers, a likelihood of a semantic concept being depicted in a frame of the video and assigning a label corresponding to the semantic concept to the frame of the video based on the likelihood of the semantic concept being depicted in the frame of the video;
selecting a plurality of representative frames of the video, wherein each representative frame is selected based on determining that the assigned label is relevant to the received search query; and
causing a player interface to be presented, wherein the video is presented within the player interface, wherein the player interface includes a progress bar that indicates a length of the video, wherein a marker is presented on the progress bar that indicates a time in the video at which a representative frame of the plurality of representative frames of the video occurs, wherein the representative frame is selected from among the plurality of representative frames based on relevance to the received search query, and wherein interaction with the marker causes the representative frame to be presented adjacent to the marker within the progress bar along with a textual description of the semantic concept being depicted in the representative frame.