Method and system for item identification
The method for item identification preferably includes determining visual information for an item; calculating a first encoding using the visual information; calculating a second encoding using the first encoding; determining an item identifier for the item using the second encoding; optionally presenting information associated with the item to a user; and optionally registering a new item.
1. A system, comprising:
a central system; and
a set of local systems communicatively connected to the central system, wherein each local system comprises:
a local item repository;
an imaging system configured to capture an image of an item; and
a processing system configured to:
determine an item encoding for the item based on the image, using a model trained at the central system;
receive item information comprising an item identifier; and
store the item encoding in association with the item information in the local item repository.
2. The system of claim 1 , wherein each local system is a kiosk.
3. The system of claim 1 , wherein, for each local system, the imaging system is further configured to capture an image of a second item, wherein the processing system is further configured to:
determine a second item encoding based on the image of the second item, using the model; and
determine an item identifier for the second item based on a comparison between the second item encoding and the item encoding.
4. The system of claim 1 , wherein the item encoding is output by an intermediate layer of the model.
5. The system of claim 1 , wherein the central system comprises a cloud computing system, remote from each local system.
6. The system of claim 1 , wherein the processing system is further configured to transmit the image and the item information to the central system, wherein the model is retrained at the central system based on the image.
7. The system of claim 6 , wherein the central system comprises a central item repository storing a set of predetermined encodings, wherein an updated set of predetermined encodings is determined using the retrained model, wherein the local item repository at each local system is updated based on the updated set of predetermined encodings.
8. The system of claim 7 , wherein updating the local item repository at each local system comprises: transmitting a difference between a prior set of predetermined encodings and the updated set of predetermined encodings to each local system; and updating the local item repository at each local system based on the difference.
9. The system of claim 1 , wherein the item information further comprises at least one of: an item price or an item numerosity.
10. The system of claim 1 , wherein the item comprises a food plate.
11. The system of claim 1 , wherein the imaging system comprises a camera.
12. A system, comprising:
a central system comprising a central database storing a set of predetermined encodings for a set of items; and
a set of local systems communicatively connected to the central system, wherein each local system is configured to:
capture an image of an item;
determine an item encoding based on the image, using a trained model; and
determine an item identifier for the item based on the item encoding and the set of predetermined encodings.
13. The system of claim 12 , wherein each local system comprises a local database storing the set of predetermined encodings, wherein determining the item identifier comprises accessing the set of predetermined encodings from the local database.
14. The system of claim 12 , wherein each local system is further configured to:
capture a new image for a new item;
receive a new item identifier for the new item;
transmit the new image and the new item identifier to the central system;
determine a new item encoding based on the new image; and
update the central database based on the new item encoding and the new item identifier, wherein the new item encoding is subsequently used to identify the new item.
15. The system of claim 12 , wherein the central system comprises a cloud computing system, remote from each local system.
16. The system of claim 12 , wherein the item comprises a food plate.
17. The system of claim 12 , wherein determining the item identifier comprises determining the item identifier based on a similarity between the item encoding and each of the set of predetermined encodings.
18. The system of claim 12 , wherein the trained model comprises an item classifier and a combination classifier, wherein determining the item encoding comprises:
determining a first encoding from an intermediate layer of the item classifier based on the image; and
determining a second encoding from an intermediate layer of the combination classifier based on the first encoding.
19. The system of claim 12 , wherein each local system is further configured to facilitate purchase of the item based on the item identifier.
20. The system of claim 12 , wherein each local system is a kiosk.