Augmented-reality-interface conflation identification
A method for detecting conflation errors in an augmented reality interface is disclosed. In one embodiment, such a method includes receiving an actual screenshot of an augmented reality interface. The method generates, using a generative adversarial network (GAN), a replica screenshot of the augmented reality interface that indicates how the actual screenshot is expected to appear. The replica screenshot is non-identical to the actual screenshot. The method receives the actual screenshot and the replica screenshot into a fractal-based convolutional neural network (CNN) to determine a correlation between the actual screenshot and the replica screenshot. In the event the correlation is below a designated threshold, the method flags the actual screenshot as having a potential conflation error. This conflation error may involve conflating a digital component for a physical component or vice versa. A corresponding system and computer program product are also disclosed.
1 . A method for detecting conflation errors in an augmented reality interface, the method comprising:
dividing the augmented reality interface into physical components and digital components to generate a set of appended encodings, wherein the generating the set of appended encodings comprises:
generating time-stamped physical and digital foundation model encodings based on, respectively, representations of the physical components and the digital components in the augmented reality interface;
generating time-stamped migration encodings based on migrating and/or exchanging the physical components and the digital components within the augmented reality interface; and
generating time-stamped ablation encodings based on ablating the physical components and the digital components one at a time;
applying a conflation error classifier neural network to the set of appended encodings;
receiving an actual screenshot of the augmented reality interface;
generating, using a generative adversarial network (GAN) and based on inputs comprising the representations of the physical components and the digital components, a replica screenshot of the augmented reality interface indicating how the actual screenshot is expected to appear;
determining, via a fractal-based convolutional neural network (CNN), a correlation between the actual screenshot and the replica screenshot;
flagging the actual screenshot as having a potential conflation error in response to determining that the correlation is below a designated threshold; and
determining a confidence of the potential conflation error based on a result of the applying the conflation error classifier.
2 . The method of claim 1 , wherein the conflation error involves conflating, in the augmented reality interface, a digital component for a physical component.
3 . The method of claim 1 , wherein the conflation error involves conflating, in the augmented reality interface, a physical component for a digital component.
4 . The method of claim 1 , wherein the replica screenshot indicates how the augmented reality interface is expected to appear when taking into account characteristics of a device on which the augmented reality interface is presented.
5 . The method of claim 4 , wherein the characteristics include at least one of screen size, language, notch dimensions, and operating system.
6 . The method of claim 1 , wherein the replica screenshot indicates how the augmented reality interface is expected to appear when taking into account design characteristics of the augmented reality interface.
7 . The method of claim 1 , wherein the correlation is represented as a percent correlation.
8 . A computer program product for detecting conflation errors in an augmented reality interface, the computer program product comprising a computer-readable storage medium having computer-usable program code embodied therein, the computer-usable program code configured to perform the following when executed by at least one processor:
divide the augmented reality interface into physical components and digital components to generate a set of appended encodings, wherein the generating the set of appended encodings comprises:
generating time-stamped physical and digital foundation model encodings based on, respectively, representations of the physical components and the digital components in the augmented reality interface;
generating time-stamped migration encodings based on migrating and/or exchanging the physical components and the digital components within the augmented reality interface; and
generating time-stamped ablation encodings based on ablating the physical components and the digital components one at a time;
apply a conflation error classifier neural network to the set of appended encodings;
receive an actual screenshot of the augmented reality interface;
generate, using a generative adversarial network (GAN) and based on inputs comprising the representations of the physical components and the digital components, a replica screenshot of the augmented reality interface indicating how the actual screenshot is expected to appear;
determine, via a fractal-based convolutional neural network (CNN), a correlation between the actual screenshot and the replica screenshot;
flag the actual screenshot as having a potential conflation error in response to determining that the correlation is below a designated threshold; and
determine a confidence of the potential conflation error based on a result of the applying the conflation error classifier.
9 . The computer program product of claim 8 , wherein the conflation error involves conflating, in the augmented reality interface, a digital component for a physical component.
10 . The computer program product of claim 8 , wherein the conflation error involves conflating, in the augmented reality interface, a physical component for a digital component.
11 . The computer program product of claim 8 , wherein the replica screenshot indicates how the augmented reality interface is expected to appear when taking into account characteristics of a device on which the augmented reality interface is presented.
12 . The computer program product of claim 11 , wherein the characteristics include at least one of screen size, language, notch dimensions, and operating system.
13 . The computer program product of claim 8 , wherein the replica screenshot indicates how the augmented reality interface is expected to appear when taking into account design characteristics of the augmented reality interface.
14 . The computer program product of claim 8 , wherein the correlation is represented as a percent correlation.
15 . A system for detecting conflation errors in an augmented reality interface, the system comprising:
at least one processor; and
at least one memory device operably coupled to the at least one processor and storing instructions for execution on the at least one processor, the instructions causing the at least one processor to:
divide the augmented reality interface into physical components and digital components to generate a set of appended encodings, wherein the generating the set of appended encodings comprises:
generating time-stamped physical and digital foundation model encodings based on, respectively, representations of the physical components and the digital components in the augmented reality interface;
generating time-stamped migration encodings based on migrating and/or exchanging the physical components and the digital components within the augmented reality interface; and
generating time-stamped ablation encodings based on ablating the physical components and the digital components one at a time;
apply a conflation error classifier neural network to the set of appended encodings;
receive an actual screenshot of the augmented reality interface;
generate, using a generative adversarial network (GAN) and based on inputs comprising the representations of the physical components and the digital components, a replica screenshot of the augmented reality interface indicating how the actual screenshot is expected to appear;
determine, via a fractal-based convolutional neural network (CNN), a correlation between the actual screenshot and the replica screenshot;
flag the actual screenshot as having a potential conflation error in response to determining that the correlation is below a designated threshold; and
determine a confidence of the potential conflation error based on a result of the applying the conflation error classifier.
16 . The system of claim 15 , wherein the conflation error involves conflating, in the augmented reality interface, a digital component for a physical component.
17 . The system of claim 15 , wherein the conflation error involves conflating, in the augmented reality interface, a physical component for a digital component.
18 . The system of claim 15 , wherein the replica screenshot indicates how the augmented reality interface is expected to appear when taking into account characteristics of a device on which the augmented reality interface is presented.
19 . The system of claim 18 , wherein the characteristics include at least one of screen size, language, notch dimensions, and operating system.
20 . The system of claim 15 , wherein the replica screenshot indicates how the augmented reality interface is expected to appear when taking into account design characteristics of the augmented reality interface.