Methods for object recognition and related arrangements

Abstract

Methods and arrangements involving portable user devices such smartphones and wearable electronic devices are disclosed, as well as other devices and sensors distributed within an ambient environment. Some arrangements enable a user to perform an object recognition process in a computationally- and time-efficient manner. Other arrangements enable users and other entities to, either individually or cooperatively, register or enroll physical objects into one or more object registries on which an object recognition process can be performed. Still other arrangements enable users and other entities to, either individually or cooperatively, associate registered or enrolled objects with one or more items of metadata. A great variety of other features and arrangements are also detailed.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS[0001]This application claims the benefit of U.S. Provisional Application No. 61 / 811,049, filed Apr. 11, 2013, U.S. Provisional Application No. 61 / 815,172, filed Apr. 23, 2013, U.S. Provisional Application No. 61 / 815,994, filed Apr. 25, 2013 and U.S. Provisional Application No. 61 / 838,165, filed Jun. 21, 2013, each of which is herein incorporated by reference.TECHNICAL FIELD[0002]The present technology generally concerns object recognition techniques, search space reduction techniques, processing techniques that may be described as contextual, anticipatory or intuitive, techniques for implementing location-based services, ubiquitous or crowd-sourced capture of imagery, sound or other data to support the above-mentioned techniques, and many other technologies.BACKGROUND AND SUMMARY[0003]Image fingerprinting (aka image signature technology) commonly involves deriving a set of 2D feature points from imagery, and then searching a set of reference i...

AI Technical Summary

Benefits of technology

[0014]4. Device, Local-Server, Global-Server Dynamics: Many retail, in-store applications will push key reference features directly onto the user's device, allowing fast device-side execution of object recognition, constrained by power consumption, memory and channel usage. Fluidity of “where” various recognition stages are actually executed provides a welcome design flexibility in the device-local-global continuum.

Problems solved by technology

But they break down when trying identifying 3D objects.

But if the camera view is oblique, as in FIG. 1, conventional fingerprinting starts having difficulty.

Even if a reference fingerprint (e.g., SIFT) was available for the entire package (e.g., in a flat configuration, before the box was glued into its 3D configuration), this reference fingerprint will be difficult to match with the FIG. 1, given its projective distortions in two opposing directions.

Still more difficult are objects that cannot readily be fingerprinted in a “flat” state (e.g., before a box is glued).