Methods to present search keywords for image-based queries
Techniques for providing recommended keywords in response to an image-based query are disclosed herein. In particular, various embodiments utilize an image matching service to identify recommended search keywords associated with image data received from a user. The search keywords can be used to perform a keyword search to identify content associated with an image input that may be relevant. For example, an image search query can be received from a user. The image search query may result in multiple different types of content that are associated with the image. The system may present keywords associated with matching images to allow a user to further refine their search and/or find other related products that may not match with the particular image. This enables users to quickly refine a search using keywords that may be difficult to identify otherwise and to find the most relevant content for the user.
1. A computer-implemented method, comprising:
identifying an object of interest in multimedia data;
identifying, based at least in part on at least one feature of the object of interest in the multimedia data, a first plurality of content items respectively associated with information related to one or more products;
removing at least one content item from the first plurality of content items;
identifying a second plurality of content items from the remaining content items, the second plurality of content items associated with a plurality of keywords;
identifying, from the plurality of keywords, a subset including one or more unique keywords;
generating, based at least in part on the subset of the plurality of keywords, a search string including at least one keyword of the subset of keywords, the search string including the at least one keyword configured for selection on a client computing device;
causing the search string and one or more items of the second plurality of content items to be displayed by the client computing device; and
receiving a keyword query from the client computing device in response to a selection of the search string on the client computing device;
searching at least one data store of content items by comparing content within the second plurality of content items to the at least one keyword of the plurality of keywords;
identifying a subset of content items matching the at least one keyword; and
causing the subset of content items to be displayed by the client computing device.
2. The computer-implemented method of claim 1 , wherein the multimedia data includes an image, a video, or an audio recording.
3. The computer-implemented method of claim 1 , further comprising:
generating a second search string including at least one keyword differing from the at least one keyword in the search string; and
causing the second search string to be displayed by the client computing device.
4. The computer-implemented method of claim 3 , wherein before providing the first search string or the second search string, the method further comprises:
identifying a priority of the first search string and a priority of the second search string; and
causing the first search string and the second search string to be displayed, by the client computing device, in an order of highest to lowest priority.
5. The computer-implemented method of claim 1 , wherein identifying the first plurality of content items associated with the multimedia data, further comprises:
obtaining a similarity confidence score for each of the first plurality of content items; and
removing, from the first plurality of content items, content items below a predetermined similarity confidence score threshold.
6. The computer-implemented method of claim 1 , wherein identifying an object of interest in multimedia data further comprises:
searching at least one data store by comparing the at least one feature of the object of interest to one or more stored features associated with a plurality of multimedia in the at least one data store; and
identifying a subset of the plurality of multimedia having features matching the at least one feature of the object of interest, wherein each of the subset of the plurality of multimedia is associated with one of the first plurality of content items.
7. The computer-implemented method of claim 1 , further comprising:
causing a subset of content items to be displayed by the client computing device in response to a selection of the generated search string.
8. The computer-implemented method of claim 1 , wherein generating the search string further comprises:
identifying an occurrence rate of each keyword within the plurality of keywords; and
selecting one or more of the keywords with the highest occurrence rate within the plurality of keywords.
9. The computer-implemented method of claim 1 , wherein generating a search string further comprises:
identifying relevance scores for each keyword in the plurality of keywords, the relevance scores indicating a discriminative value of each keyword based on the first plurality of content items; and
selecting the at least one keyword as a search link based on the relevance scores for each keyword in the plurality of keywords.
10. A computing system, comprising:
at least one processor; and
a memory device including instructions that, when executed by the at least one processor, cause the computing system to:
obtain an image of an item;
transmit the image of the item to a server computer in an image query, wherein the server computer is configured to:
identify an object of interest in multimedia data;
identify, based at least in part on at least one feature of the object of interest in the multimedia data, a first plurality of content items respectively associated with information related to one or more products;
remove at least one content item from the first plurality of content items;
identify a second plurality of content items from the remaining content items, the second plurality of content items associated with a plurality of keywords;
identify, from the plurality of keywords, a subset including one or more unique keywords; and
generate, based at least in part on the subset of the plurality of keywords, a search string including at least one keyword of the subset of keywords, the search string including the at least one keyword configured for selection on a client computing device; and
cause the search string and one or more items of the second plurality of content items to be displayed by the client computing device;
receive a keyword query from the client computing device in response to a selection of the search string on the client computing device;
search at least one data store of content items by comparing content within the second plurality of content items to the at least one keyword of the plurality of keywords;
identify a subset of content items matching the at least one keyword; and
cause the subset of content items to be displayed by the client computing device.
11. The computing system of claim 10 , wherein the multimedia data includes an image, a video, or an audio recording.
12. The computing system of claim 10 , wherein the instructions, when executed by the at least one processor, further enable the server computer to:
generate a second search string including at least one keyword differing from the at least one keyword in the search string; and
cause the second search string to be displayed by the client computing device.
13. The computing system of claim 12 , wherein the instructions, when executed by the at least one processor, further enable the server computer to, before providing the first search string or the second search string:
identify a priority of the first search string and a priority of the second search string; and
cause the first search string and the second search string to be displayed, by the client computing device, in an order of highest to lowest priority.
14. A non-transitory computer readable storage medium storing instructions that, when executed by at least one processor of a server computer, cause the server computer to:
identify an object of interest in multimedia data;
identify, based at least in part on at least one feature of the object of interest in the multimedia data, a first plurality of content items respectively associated with information related to one or more products;
remove at least one content item from the first plurality of content items;
identify a second plurality of content items from the remaining content items, the second plurality of content items associated with a plurality of keywords;
identify, from the plurality of keywords, a subset including one or more unique keywords;
generate, based at least in part on the subset of the plurality of keywords, a search string including at least one keyword of the subset of keywords, the search string including the at least one keyword configured for selection on a client computing device;
cause the search string and one or more items of the second plurality of content items to be displayed by the client computing device; and
receive a keyword query from the client computing device in response to a selection of the search string on the client computing device;
search at least one data store of content items by comparing content within the second plurality of content items to the at least one keyword of the plurality of keywords;
identify a subset of content items matching the at least one keyword; and
cause the subset of content items to be displayed by the client computing device.
15. The non-transitory computer readable storage medium of claim 14 , wherein the instructions, when executed by the at least one processor, further enable the server computer to:
obtain a similarity confidence score for each of the first plurality of content items; and
remove, from the first plurality of content items, content items below a predetermined similarity confidence score threshold.
16. The non-transitory computer readable storage medium of claim 14 , wherein the instructions, when executed by the at least one processor, further enable the server computer to:
cause a subset of content items to be displayed by the client computing device in response to a selection of the generated search string.
17. The non-transitory computer readable storage medium of claim 14 , wherein the instructions, when executed by the at least one processor, further enable the server computer to:
identify an occurrence rate of each keyword within the plurality of keywords; and
select one or more of the keywords with the highest occurrence rate within the plurality of keywords.
18. The non-transitory computer readable storage medium of claim 14 , wherein the instructions, when executed by the at least one processor, further enable the server computer to:
identify relevance scores for each keyword in the plurality of keywords, the relevance scores indicating a discriminative value of each keyword based on the second plurality of content items; and
select the at least one keyword as a search link based on the relevance scores for each keyword in the plurality of keywords.