Generating and utilizing normalized scores for classifying digital objects
First Claim
Patent Images
1. A method comprising:
- generating, by at least one processor and utilizing a model trained to identify a known object based on a set of digital training images portraying the known object, a classification score with regard to an unknown object portrayed in a probe digital image, the classification score indicating a likelihood that the unknown object portrayed in the probe digital image corresponds to the known object portrayed in the set of digital training images;
using a probability function reflecting probabilities of returning classification scores, wherein the probability function is specific to the number of digital training images in the set of digital training images, to transform the classification score into a normalized classification score; and
determining, using the normalized classification score, whether the unknown object portrayed in the probe digital image corresponds to the known object portrayed in the set of digital training images.
2 Assignments
0 Petitions
Accused Products
Abstract
The present disclosure is directed toward systems and methods that enable more accurate digital object classification. In particular, disclosed systems and methods address inaccuracies in digital object classification introduced by variations in classification scores. Specifically, in one or more embodiments, disclosed systems and methods generate probability functions utilizing digital test objects and transform classifications scores into normalized classification scores utilizing probability functions. Disclosed systems and methods utilize normalized classification scores to more accurately classify and identify digital objects in a variety of applications.
22 Citations
20 Claims
-
1. A method comprising:
-
generating, by at least one processor and utilizing a model trained to identify a known object based on a set of digital training images portraying the known object, a classification score with regard to an unknown object portrayed in a probe digital image, the classification score indicating a likelihood that the unknown object portrayed in the probe digital image corresponds to the known object portrayed in the set of digital training images; using a probability function reflecting probabilities of returning classification scores, wherein the probability function is specific to the number of digital training images in the set of digital training images, to transform the classification score into a normalized classification score; and determining, using the normalized classification score, whether the unknown object portrayed in the probe digital image corresponds to the known object portrayed in the set of digital training images. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8)
-
-
9. A system comprising:
-
at least one processor; and at least one non-transitory computer readable storage medium storing instructions that, when executed by the at least one processor, cause the system to; generate a classification score with regard to an unknown object portrayed in a probe digital image based on a set of tagged digital training images portraying a known object, the classification score indicating a likelihood that the unknown object portrayed in the probe digital image corresponds to the known object portrayed in the set of tagged digital training images; transform the classification score into a normalized classification score based on a probability function reflecting probabilities of returning classification scores, wherein the probability function is specific to the number of tagged digital training images in the set of tagged digital training images; and determine, using the normalized classification score, that the unknown object portrayed in the probe digital image corresponds to the known object portrayed in the set of tagged digital training images. - View Dependent Claims (10, 11, 12, 13, 14)
-
-
15. A non-transitory computer readable medium storing instruction thereon that, when executed by at least one processor, cause a computer system to:
-
generate a classification score with regard to an unknown object portrayed in a probe digital image based on a set of tagged digital training images portraying a known object, the classification score indicating a likelihood that the unknown object portrayed in the probe digital image corresponds to the known object portrayed in the set of tagged digital training images; transform the classification score into a normalized classification score based on a probability function reflecting probabilities of returning classification scores, wherein the probability function is specific to the number of tagged digital training images in the set of tagged digital training images; and determine, using the normalized classification score, that the unknown object portrayed in the probe digital image corresponds to the known object portrayed in the set of tagged digital training images. - View Dependent Claims (16, 17, 18, 19, 20)
-
Specification