Video segments for a video related to a task
Methods and apparatus related to identifying a video for completing a task and determining a plurality of video segments of the identified video based on one or more attributes of the task. A task and a plurality of how-to videos related to the task may be identified. A how-to video may be selected and a plurality of video segments of the selected how-to video may be determined. One or more video segments may be associated with one or more task attributes that relate to performing the task. The selected video may be provided to a user and segmented, indexed, and/or annotated based on the associated video segments. In some implementations a given object utilized in performing the task may be identified and one or more video segments corresponding to the given object may be identified and/or provided to the user.
1. A method implemented by one or more processors, the method comprising:
identifying a plurality of how-to videos for a task;
determining a corresponding confidence measure for each of the plurality of how-to videos;
selecting a how-to video, of the plurality of how-to videos, based on the corresponding confidence measure for the how-to video;
identifying one or more task attributes of the selected how-to video for the task;
determining a plurality of video segments of the selected how-to video by segmenting the selected how-to video based on a transcript of the selected how-to video, wherein determining the plurality of video segments comprises:
determining a given segment, of the plurality of the video segments, based on matching terms, of the transcript of the given segment, to the identified one or more task attributes of the selected how-to video for the task;
storing an association of the given segment to the identified one or more task attributes of the selected how-to video based on the matching terms;
subsequent to the storing:
receiving a query;
determining that the given segment is responsive to the query; and
in response to determining that the given segment is responsive to the query:
providing, in response to the query, a link to the given segment of the how-to video.
2. The method of claim 1 , wherein the query is a spoken query.
3. The method of claim 1 , wherein the query comprises an image, and wherein determining that the given segment is responsive to the query comprises matching one or more objects, detected in the image, to one or more of the task attributes.
4. The method of claim 1 , wherein selecting the how-to video based on the corresponding confidence measure for the how-to video comprises selecting the how-to video based on the corresponding confidence measure satisfying a threshold.
5. The method of claim 1 , wherein determining the corresponding confidence measure for the how-to video is based on a measure of popularity of the how-to video.
6. The method of claim 1 , further comprising:
providing, in response to the query, a visual indication of the corresponding confidence measure of the how-to video.
7. A system, comprising:
a database that stores an association of a given segment, out of a plurality of segments that are segmented from a how-to video for performing a task, to a given object utilized in performing the task;
memory storing instructions; and
one or more processors operable to execute the instructions stored in the memory, wherein the instructions comprise instructions to:
determine the given segment, out of the plurality of segments that are segmented from the how-to video for performing the task, based on matching terms, of a transcript of the given segment, to the given object utilized in performing the task;
store, in the database, the association of the given segment to the given object based on the matching terms;
receive, from a client device, at least one image that captures one or more objects, including the given object;
process the at least one image to identify the given object;
determining that the given segment is responsive to the at least one image based on the stored association of the given segment to the given object identified from processing the at least one image;
in response to determining that the given segment is responsive to the at least one image:
providing a link to the given segment of the how-to video.
8. The system of claim 7 , wherein the instructions further comprise instructions to:
identify a plurality of how-to videos for the task;
determine a corresponding confidence measure for each of the how-to videos;
select the how-to video, of the how-to videos, based on the corresponding confidence measure for the how-to video; and
store, in the database based on the corresponding confidence measure for the how-to video, the association of the given segment to the given object.
9. The system of claim 8 , wherein the instructions to select the how-to video based on the corresponding confidence measure for the how-to video comprise instructions to select the how-to video based on the corresponding confidence measure satisfying a threshold.
10. The system of claim 9 , wherein the instructions to determine the corresponding confidence measure for the how-to video comprise instructions to determine the corresponding confidence measure based on a measure of popularity of the how-to video.
11. The system of claim 8 , wherein the instructions further comprise instructions to:
provide, along with the link to the given segment of the how-to-video, a visual indication of the corresponding confidence measure of the how-to video.
12. One or more non-transitory computer-readable media storing instructions that, when executed by one or more processors, cause the one or more processors to perform a method comprising:
identifying a plurality of how-to videos for a task;
determining a corresponding confidence measure for each of the plurality of how-to videos;
selecting a how-to video, of the plurality of how-to videos, based on the corresponding confidence measure for the how-to video;
identifying one or more task attributes of the selected how-to video for the task;
determining a plurality of video segments of the selected how-to video by segmenting the selected how-to video based on a transcript of the selected how-to video, wherein determining the plurality of video segments comprises:
determining a given segment, of the plurality of video segments, based on matching terms, of the transcript of the given segment, to the identified one or more task attributes of the selected how-to video for the task;
storing an association of the given segment to the identified one or more task attributes of the selected how-to video based on the matching terms;
subsequent to the storing;
receiving a query;
determining that the given segment is responsive to the query; and
in response to determining that the given segment is responsive to the query:
providing, in response to the query, a link to the given segment of the how-to video.
13. The one or more non-transitory computer-readable media of claim 12 , wherein the query is a spoken query.
14. The one or more non-transitory computer-readable media of claim 12 , wherein the query comprises an image, and wherein determining that the given segment is responsive to the query comprises matching one or more objects, detected in the image, to one or more of the task attributes.
15. The one or more non-transitory computer-readable media of claim 12 , wherein selecting the how-to video based on the corresponding confidence measure for the how-to video comprises selecting the how-to video based on the corresponding confidence measure satisfying a threshold.
16. The one or more non-transitory computer-readable media of claim 15 , wherein determining the corresponding confidence measure for the how-to video is based on a measure of popularity of the how-to video.