Low-overhead image search result generation
View Patent ↗A device may be configured to identify a plurality of images that are similar to a query image; generate a plurality of sets of rankings of the identified images based on a plurality of image attributes; compare the generated plurality of sets of rankings of the identified images to a reference set of rankings of images; select, based on the comparing, a particular set of rankings; and rank a plurality of images that are associated with another query image, based on an attribute associated with the selected particular set of rankings.
1. A method comprising:
generating, by a device and based on a first plurality of images grouped into a plurality of clusters, a ranking of the first plurality of images,
each cluster, of the plurality of clusters, including at least one of the first plurality of images, and
the ranking of the first plurality of images being based on a respective image attribute associated with each image, of the first plurality of images, in each cluster of the plurality of clusters;
comparing, by the device, the ranking with a set of rankings of images;
selecting, by the device and based on the comparing, a particular set of rankings from the set of the rankings of the images; and
providing, by the device and for presentation, a second plurality of images,
the second plurality of images being ranked based on an attribute associated with the selected particular set of rankings.
2. The method of claim 1 , further comprising:
generating the set of the rankings of the images.
3. The method of claim 2 , where, when generating the set of the rankings of the images, the method includes:
generating the set of the rankings of the images based on applying different attributes to each set of the rankings of the images.
4. The method of claim 1 , where, when selecting the particular set of the rankings, the method includes:
comparing an order of images of the first plurality of images with an order of images in each set of the rankings of the images; and
selecting, based on comparing the order of images of the first plurality of images with the order of images in each set of the rankings of the images, the particular set of the rankings.
5. The method of claim 1 , where the first plurality of images are grouped into the plurality of clusters based on receiving a query image.
6. The method of claim 5 , further comprising:
receiving another query image;
determining that the other query image is similar to the query image; and
ranking, based on determining that the other query image is similar to the query image, a plurality of images, associated with the other query image, based on the attribute associated with the selected particular set of rankings.
7. The method of claim 1 , where the attribute, associated with the selected particular set of rankings, corresponds to dimension information.
8. A system comprising:
one or more devices to:
generate, based on a first plurality of images grouped into a plurality of clusters, a ranking of the first plurality of images,
each cluster, of the plurality of clusters, including at least one of the first plurality of images, and
the ranking of the first plurality of images being based on a respective image attribute associated with each image, of the first plurality of images, in each cluster of the plurality of clusters;
compare the ranking with a set of rankings of images;
select, based on the comparing, a particular set of rankings from the set of the rankings of the images; and
provide, for presentation, a second plurality of images,
the second plurality of images being ranked based on an attribute associated with the selected particular set of rankings.
9. The system of claim 8 , where the one or more devices are further to:
generate the set of the rankings of the images.
10. The system of claim 9 , where the one or more devices, when generating the set of the rankings of the images, are further to:
generate the set of the rankings of the images based on applying different attributes to each set of the rankings of the images.
11. The system of claim 8 , where the one or more devices, when selecting the particular set of the rankings, are further to:
compare an order of images of the first plurality of images with an order of images in each set of the rankings of the images; and
select, based on comparing the order of images of the first plurality of images with the order of images in each set of the rankings of the images, the particular set of the rankings.
12. The system of claim 8 , where the first plurality of images are grouped into the plurality of clusters based on receiving a query image.
13. The system of claim 12 , where the one or more devices are further to:
receive another query image;
determine that the other query image is similar to the query image; and
rank, based on determining that the other query image is similar to the query image, a plurality of images, associated with the other query image, based on the attribute associated with the selected particular set of rankings.
14. The system of claim 8 , where the attribute, associated with the selected particular set of rankings, corresponds to dimension information.
15. A non-transitory computer-readable medium storing instructions, the instructions comprising:
one or more instructions which, when executed by at least one processor, cause the at least one processor to:
generate, based on a first plurality of images grouped into a plurality of clusters, a ranking of the first plurality of images,
each cluster, of the plurality of clusters, including at least one of the first plurality of images, and
the ranking of the first plurality of images being based on a respective image attribute associated with each image, of the first plurality of images, in each cluster of the plurality of clusters;
compare the ranking with a set of rankings of images;
select, based on the comparing, a particular set of rankings from the set of the rankings of the images; and
provide, for presentation, a second plurality of images,
the second plurality of images being ranked based on an attribute associated with the selected particular set of rankings.
16. The non-transitory computer-readable medium of claim 15 , where the one or more instructions further include:
one or more instructions to generate the set of the rankings of the images.
17. The non-transitory computer-readable medium of claim 16 , where the one or more instructions to generate the set of the rankings of the images include:
one or more instructions to generate the set of the rankings of the images based on applying different attributes to each set of the rankings of the images.
18. The non-transitory computer-readable medium of claim 15 , where the one or more instructions to select the particular set of the rankings include:
one or more instructions to compare an order of images of the first plurality of images with an order of images in each set of the rankings of the images; and
one or more instructions to select, based on comparing the order of images of the first plurality of images with the order of images in each set of the rankings of the images, the particular set of the rankings.
19. The non-transitory computer-readable medium of claim 15 , where the first plurality of images are grouped into the plurality of clusters based on receiving a query image.
20. The non-transitory computer-readable medium of claim 19 , where the instructions further include:
one or more instructions to receive another query image;
one or more instructions to determine that the other query image is similar to the query image; and
one or more instructions to rank, based on determining that the other query image is similar to the query image, a plurality of images, associated with the other query image, based on the attribute associated with the selected particular set of rankings.