Object search and navigation method and system

摘要

A computer implemented method of presenting visual content items, comprising: grouping the visual content items according to predefined similarity rules relating to visual characteristics of the visual content items such that each group has a range for the number of its members; selecting a representative visual content item for each group; presenting the representative visual content item of each group that has a minimal number of members above a predefined threshold; and optionally presenting the visual content items alongside the representative visual content items. The grouping may be carried out by using predefined color groups; by using predefined shape groups; or using Keypoints of the visual content items, and relate to groups of human faces, product images and landscape images.

主权项

1. A computer implemented method of presenting a plurality of visual content items, comprising:
receiving a text query submitted by a user; grouping a plurality of visual content items which are provided as query results to said text query in a plurality of groups according to a plurality of predefined similarity rules relating to a plurality of visual characteristics of the plurality of visual content items; selecting a plurality of representative visual content items each for one group of the plurality of groups, each one of said plurality of representative visual content-items is selected according to at least one image parameter thereof, each one of said plurality of representative visual content items is selected from the plurality of visual content items of a respective said group; presenting a graphical user interface having:
a first subpart presenting said plurality of representative visual content items each of one of the plurality of groups;a second subpart presenting a plurality of categorization items each indicative of a different color;a third subpart presenting members of one of the plurality of groups; and filtering at least some of said plurality of groups or at least some of said plurality of visual content items in said plurality of groups in response to at least one user selection of at least one of said categorization items in said second subpart; updating a presentation of said graphical user interface on a display of a user terminal in response to an outcome of said filtering; labeling each group of the plurality of groups with a group label selected from a plurality of text fields of members of the respective group; wherein the group label describes at least one non visual characteristic of the respective representative visual content item; wherein said presentation comprises presenting said group label; wherein the plurality of visual content items are images of a plurality of products and the least one non visual characteristic in indicative of a member of a group consisting of; who made a product, when a product was made, where a product was made, a content of a product, a product usage instructions, and a safely of a product.