Feature-based search
Various embodiments of systems and methods allow a system to identify subsets of items by mixing and matching identified features in one or more other items. A system can identify features of items in an item database. The system can then calculate “fingerprints” of these features which are vectors describing the characteristics of the features. The system can present a collection of items and a user can select an item of the collection. The user can then select positive features to include in a search and/or negative features to include in the search. The system can then do a search of the database for items that contain features similar to those positive features and do not contain features similar to those negative features. The user can select features through a variety of means.
1. A computer-implemented method for searching a collection of items, comprising:
displaying a plurality of shoes of a collection of shoes;
receiving a selection of a first shoe of the collection of shoes;
identifying one or more features of the first shoe based on a first image for the first shoe;
receiving a selection of a first feature of the one or more features;
calculating a first fingerprint of the first feature using a first portion of the first image, the first portion corresponding to the first feature;
receiving a selection of a second feature of a second shoe of the collection of shoes;
calculating a second fingerprint of the second feature using a second portion of a second image, the second portion corresponding to the second feature; and
determining a subset of the collection of shoes that have respective features similar to the first and second features, based on the first fingerprint and the second fingerprint.
2. The computer-implemented method of claim 1 , further comprising:
presenting an indication on the first image or the second image that the at least one feature is selectable.
3. The computer-implemented method of claim 1 , further comprising:
receiving a selection of a third feature; and
calculating a third fingerprint of the third feature, determining that the subset of the collection of shoes do not have respective features similar to the third feature, based on the third fingerprint.
4. A computer-implemented method comprising:
obtaining a collection of items;
identifying one or more features of an item of the collection of items based on an image for the item;
presenting the image for the item;
receiving a selection of at least one feature of the one or more features;
calculating a fingerprint of the at least one feature;
determining a subset of the collection of items that have respective features similar to the at least one feature, based on the fingerprint; and
presenting the subset of the collection of items.
5. The computer-implemented method of claim 4 , further comprising:
receiving a selection of a second feature of a second item of the collection of items;
calculating a second fingerprint of the second feature; and
determining a second subset of the subset of the collection of items that have respective features similar to the second feature, based on the second fingerprint.
6. The computer-implemented method of claim 4 , wherein receiving a selection of the at least one feature further comprises:
receiving an input to paint a region of the image; and
determining that the region of the image pertains to the at least one feature.
7. The computer-implemented method of claim 4 , further comprising:
presenting an indication on the image that the at least one feature is selectable.
8. The computer-implemented method of claim 4 , wherein the fingerprint is calculated using a convolutional neural network.
9. The computer-implemented method of claim 4 , further comprising:
determining a type of the at least one feature; and
generating the fingerprint using a process configured for the type of the at least one feature.
10. The computer-implemented method of claim 4 , further comprising:
receiving a selection of a second feature of the item, the selection of the second feature indicating that the subset of items should exclude items with the second feature; and
determining the subset of the collection of items further based on a second fingerprint of the second feature.
11. The computer-implemented method of claim 4 , further comprising:
receiving, from a client device, a user image of a second feature; and
determining the subset of the collection of items further based on a second fingerprint of the second feature.
12. The computer-implemented method of claim 4 , further comprising:
receiving an input to adjust the at least one feature resulting in an adjusted feature;
calculating an adjusting fingerprint from the adjusted feature; and
determining an adjusted subset of the collection of items based on the adjusted fingerprint.
13. A system, comprising:
at least one processor; and
memory including instructions that, when executed by the at least one processor, cause the system to:
obtain a collection of items;
identify one or more features of an item of the collection of items based on an image for the items;
present the image for the item;
receive a selection of at least one feature of the one or more features;
calculate a fingerprint of the at least one feature;
determine a subset of the collection of items that have respective features similar to the at least one feature, based on the fingerprint; and
present the subset of the collection of items.
14. The system of claim 13 , wherein the instructions when executed further cause the system to:
receive a selection of a second feature of a second item of the collection of items;
calculate a second fingerprint of the second feature; and
determine a second subset of the subset of the collection of items that have respective features similar to the second feature, based on the second fingerprint.
15. The system of claim 13 , wherein the instructions that cause the system to receive a selection of the at least one feature further cause the system to:
receive an input to paint a region of the image; and
determine that the region of the image pertains to the at least one feature.
16. The system of claim 13 , wherein the instructions when executed further cause the system to:
present an indication on the image that the at least one feature is selectable.
17. The system of claim 13 , wherein the fingerprint is calculated using a convolutional neural network.
18. The system of claim 13 , wherein the instructions when executed further cause the system to:
determine a type of the at least one feature; and
generate the fingerprint using a process configured for the type of the at least one feature.
19. The system of claim 13 , wherein the instructions when executed further cause the system to:
receive a selection of a second feature of the item, the selection of the second feature indicating that the subset of items should exclude items with the second feature; and
determine the subset of the collection of items further based on a second fingerprint of the second feature.
20. The system of claim 13 , wherein the instructions when executed further cause the system to:
receive, from a client device, a user image of a second feature; and
determine the subset of the collection of items further based on a second fingerprint of the second feature.