US 12,333,740 B1
Determining a location of a shopping container within a store
Evyatar Nimrod Ben-Shtirit, Tel Aviv (IL); Shlomi Amitai, Tel Aviv (IL); Eran Menahem Kravitz, Tel Aviv (IL); Lia Eidler-Shemesh, Tel Aviv (IL); and Raz Aharon Golan, Tel Aviv (IL)
Assigned to Shopic Technologies LTD., Tel Aviv (IL)
Filed by Shopic Technologies LTD., Tel Aviv (IL)
Filed on Aug. 2, 2024, as Appl. No. 18/793,704.
Int. Cl. H04N 23/90 (2023.01); G06T 7/246 (2017.01); G06T 7/73 (2017.01)
CPC G06T 7/248 (2017.01) [G06T 7/74 (2017.01); H04N 23/90 (2023.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30244 (2013.01)] 23 Claims
OG exemplary drawing
 
1. A method for determining a location of a shopping container within a store, the method comprises:
acquiring by a first side camera associated with the shopping container, a first side image of first content located to a first side of the shopping container; wherein the first side camera has a first side camera field of view that is directed solely to content located to the first side of the shopping container;
acquiring by a second side camera associated with the shopping container, a second side image of second content located to a second side of the shopping container; wherein the second side camera has a second side camera field of view that is directed solely to content located to the second side of the shopping container; wherein the first side differs from the second side; wherein each one of the first side and the second side differ from a front of the shopping container and differs from a rear of the shopping container;
determining a pose of the first side camera, based on the first side image and by a machine learning process trained using a structure from motion model of the store, wherein the structure from motion model is generated from untagged side images acquired by side cameras of shopping containers that moved within the store during an image acquisition process, wherein the untagged side images are associated with side cameras pose information learnt during the generation of the structure from motion model;
determining a pose of the second side camera, based on the second side image and by the machine learning process trained using the structure from motion model of the store;
determining the location of the shopping container based on the pose of the first camera and on the pose of the second camera;
determining a location determination confidence level;
determining a location error when there is a mismatch that exceeds a threshold between a location estimate based on the pose of the first camera and a location estimate based on the pose of the second camera; and
capturing by a camera that differs from the first side camera and from the second side camera, one or more images of an item being inserted or removed from the shopping container.