Method of online shopping and system therefor

1 . A computer-implemented method for online shopping and order fulfillment, the method comprising:

providing a plurality of items for purchase;

receiving a purchase order for at least one item selected by a user from among the plurality of items; and

during at least part of the fulfillment of the purchase order, initiating a video conference between a user's device operated by the user and a collector's device operated by a collector;

during the video conference:

capturing a video, by the device of the collector, of the selected at least one item and one or more proximate items during discontinuous movement of the collector;

processing, using a vector embedding machine learning (ML) model, the captured video to identify the selected at least one item and the one or more proximate items based on respective identifying vectors of the selected at least one item and the one or more proximate items, wherein processing the captured video further comprises:

detecting a pause in the movement of the collector;

start the processing of the captured video in response to detecting the pause in the movement of the collector; and,

stop the processing of the captured video in response to detecting movement of the collector has resumed.

2 . The method of claim 1 , wherein the video conference occurs when at least one selected item is being picked up.

3 . The method of claim 1 , further comprising:

in response to receiving the purchase order, sharing with the user's device a video conference connection link; and

initiating the video conference via the link.

4 . The method of claim 1 , further comprising:

providing, in real-time, at least one of the following data to be displayed on the user's device at substantially the same time as the video conference:

at least part of the captured video; and

data pertaining to at least one of the proximate items.

5 . The method of claim 4 , wherein the method further comprises:

repeatedly:

adding the identified selected item to previously identified selected items, together constituting a virtual cart; and

in response to a mismatch between the virtual cart and the purchase order, taking an action.

6 . The method of claim 4 , wherein the at least part of the captured video includes video of the selected at least one item.

7 . The method of claim 4 , further comprising:

prior to providing the data, processing the captured video to identify the proximate items, wherein each identified proximate item constitutes a candidate; and

providing stored data pertaining to at least one candidate to be displayed on the user's device.

8 . The method of claim 7 comprising:

using a Machine Learning (ML) model to process the captured video.

9 . The method of claim 8 , wherein the ML model is trained to:

identify separate objects in a given video; and

classify images and text, appearing on each object, to a stored item.

10 . The method of claim 9 wherein the text includes at least one of the following data: nutrition data, name of manufacturer, list of ingredients, and allergies.

11 . The method of claim 7 , wherein processing the video further comprises calculating a likelihood score for the at least one candidate; and

wherein providing the stored data comprises providing stored data pertaining to the candidate having the highest score.

12 . The method of claim 11 , wherein the likelihood score is calculated based on at least one of the following data: history data of the user, current location of the collector, and context-based similarity of the candidate to a selected item.

13 . The method of claim 11 , wherein the likelihood score is calculated based on a degree of matching of the selected items and the candidate, to one or more stored recipes.

14 . The method of claim 11 , wherein the likelihood score is calculated based on a degree of frequency of the candidate in stored recipes.

15 . The method of claim 13 , further comprising:

providing data pertaining to at least one stored recipe associated with the candidate having the highest score.

16 . The method of claim 11 , wherein the likelihood score is calculated based on a degree of similarity of the user to other users.

17 . The method of claim 16 , wherein the degree of similarity is determined using a similarity Machine Learning (ML) model trained to classify a given user into a plurality of user classes.

18 . The method of claim 17 , wherein the similarity ML model is trained to classify a feature vector comprising a plurality of features that are extracted from a profile and activities of the given user to a plurality of user classes, wherein the activities of the user include one or more of purchases history, frequency of purchases, and history of recipes and/or items previously provided to given user.

19 . The method of claim 11 , wherein the likelihood score is calculated based on a habitual chronological order of the user.

20 . The method of claim 7 , further comprising:

receiving a selection made by the user of one of the candidates.

21 . A computer system for online shopping, the system comprising a processing circuitry comprising at least one processer and computer memory, the processing circuitry is being configured to execute a method as defined by claim 1 .

22 . A non-transitory computer readable storage medium tangibly embodying a program of instructions that, when executed by a computer, cause the computer to perform a method for online shopping as defined by claim 1 .

23 . A system of online shipping and order fulfillment, the system comprising:

a user device;

a collector device configured to communicate with the user device;

a shopping management server configured to provide a plurality of items for purchase;

receive a purchase order for at least one item selected by a user from among the plurality of items;

cause initiation of a video conference between the user device and the collector device during fulfillment of the purchase order;

cause video captured during the video conference to be processed, wherein the captured video comprises video of the selected at least one item and one or more proximate items for optional purchase;

detect a pause in the movement of the collector;

cause the captured video to be processed in response to detecting the pause in the movement of the collector; and

cause the captured video to stop being processed in response to detecting movement of the collector has resumed;

wherein, causing the video captured to be processed further comprises using a vector embedding machine learning (ML) model to identify the selected at least one item and the one or more proximate items based on respective identifying vectors of the selected at least one item and the one or more proximate items.

24 . The system of claim 23 , further comprising:

a camera located in a storage facility, configured to communicate with the shopping management server and the user's device, wherein the camera is configured to capture the captured video;

wherein the shopping management server is configured to provide, in real-time, at least one of the following data to be displayed on the user's device at substantially the same time as the video conference:

at least part of the captured video; and

data pertaining to at least one of the proximate items.

25 . The system of claim 24 , wherein the camera is configured to capture a video of the selected at least one item.

26 . The system of claim 24 , wherein the camera is connected to a collector, and wherein the camera is configured to capture the video during discontinuous movement of the collector, and wherein the shopping management server is configured to:

detect a pause in the movement of the collector;

start processing the video; and

stop processing the video in response to identifying movement.