User Authentication and Transaction Verification via Screen-Sharing of Dynamically-Changing On-Screen Content that Incorporates Machine-Transformation Encoding of User Authentication Data and Transaction Verification Data

1 . A method comprising:

while a user interacts with an electronic device to enter transaction data for an electronic transaction intended to be performed online via a remote server,

(a) activating a user-facing camera of the electronic device, and capturing a live video feed of the user while the user enters transaction data;

(b) generating locally on said electronic device an encoded on-screen visual transformation that is a machine-transformation of at least one of:

(b1) a machine-transformation of video content of one or more video frames captured in said live video feed of the user,

(b2) a machine-transformation of transaction data that were entered so far by the user via said electronic device,

(b3) a machine-transformation of user-specific biometric characteristics that were extracted from analysis of said live video feed of the user,

(b4) a machine-transformation of an image of an identification card that the user holds towards the user-facing camera of the electronic device;

(c) displaying on a screen of said electronic device: (c1) a graphical user interface having one or more fillable fields for entering transaction data, and also (c2) a dynamically-changing non-static group of pixels that depict said encoded on-screen visual transformation;

(d) continuously sharing the screen of said electronic device with a trusted remote server, which is configured to perform server-side analysis of content that the trusted remote server receives via Screen Sharing from said electronic device;

wherein the server-side analysis comprises:

(d1) decoding the dynamically-changing non-static group of pixels that depict said encoded on-screen visual transformation, as received at the trusted remote server via Screen Sharing from the electronic device of the user; wherein said decoding at the trusted remote server yields decoded information that comprises at least one of: decoded transaction data, decoded user data;

(d2) comparing the decoded data at the trusted remote server, against at least one of: (i) transaction data that the trusted remote server received from the electronic device via a communication channel other than Screen Sharing, (ii) user data that the trusted remote server received from the electronic device via a communication channel other than Screen Sharing;

(d3) based on said comparing, determining at the trusted remote server whether to block or approve a user-submitted transaction that corresponds to said transaction data.

2 . The method of claim 1 ,

wherein step (b) comprises:

generating locally on said electronic device an encoded on-screen visual transformation that is a machine-transformation of both: (b1) video content of one or more video frames captured in said live video feed of the user, and (b2) transaction data that were entered so far by the user via said electronic device.

3 . The method of claim 2 ,

wherein step (b) comprises:

(I) generating locally on said electronic device a first encoded on-screen visual transformation that is a machine-transformation via a first transformation function of video content of one or more video frames captured in said live video feed of the user;

(II) generating locally on said electronic device a second encoded on-screen visual transformation that is a machine-transformation via a second transformation function of transaction data that were entered so far by the user via said electronic device;

wherein step (c) comprises:

displaying on the screen of said electronic device: (i) the foreground layer having one or more fillable fields for entering transaction data, and also (ii) a first portion of the background layer having said first encoded on-screen visual transformation that corresponds to transformation of live video feed data, and also (iii) a second portion of the background layer having said second encoded on-screen visual transformation that corresponds to transformation of user-entered transaction data.

4 . The method of claim 3 ,

wherein the first encoded on-screen visual transformation that, is a machine-transformation via the first transformation function of video content of one or more video frames captured in said live video feed of the user, consists of a group of pixels that do not depict a human face but rather represent machine-readable data and not human-comprehensible data.

5 . The method of claim 4 ,

wherein the second encoded on-screen visual transformation, that is a machine-transformation via the second transformation function of transaction data that were entered so far by the user via said electronic device, consists of a group of pixels that do not show the transaction data in a natural language and do not show the transaction data in a human-comprehensible format but rather represent machine-readable data and not human-comprehendible data.

6 . The method of claim 3 ,

wherein two different transformation functions are executed locally on the electronic device, comprising: (i) a first transformation function that generates a first encoded on-screen visual transformation that is a machine-transformation of video content of one or more video frames captured in said live video feed of the user; and (ii) a second transformation function that generates a second encoded on-screen visual transformation that is a machine-transformation of transaction data that were entered so far by the user via said electronic device.

7 . The method of claim 3 ,

wherein a single transformation function, or a single set of transformation functions, are executed locally on the electronic device, on an aggregated input that comprises both: (i) video content of one or more video frames captured in said live video feed of the user; and (ii) transaction data that were entered so far by the user via said electronic device.

8 . The method of claim 1 , comprising:

as the user types or enters or modifies transaction data via the electronic device,

dynamically changing an on-screen machine-transformation of the transaction data that is displayed on the electronic device as the background layer and that is shared via Screen Sharing with the trusted remote server.

9 . The method of claim 8 , comprising:

as the user types or enters or modifies transaction data via the electronic device,

dynamically changing an on-screen machine-transformation of the live video feed data based on a currently-captured video frame that undergoes machine transformation into an encoded on-screen machine-transformation that is displayed on the electronic device as the background layer and that is shared via Screen Sharing with the trusted remote server.

10 . The method of claim 1 ,

wherein said encoded on-screen visual transformation comprises both:

(I) a machine-transformation of video content of one or more video frames captured in said live video feed of the user,

and also,

(II) at least one of:

(i) a machine-transformation of transaction data that were entered so far by the user via said electronic device,

(ii) a machine-transformation of user-specific biometric characteristics that were extracted from analysis of said live video feed of the user,

(iii) a machine-transformation of an image of an identification card that the user holds towards the user-facing camera of the electronic device.

11 . The method of claim 1 ,

wherein said encoded on-screen visual transformation comprises both:

(I) a machine-transformation of an image of an identification card that the user holds towards the user-facing camera of the electronic device;

and also,

(II) at least one of:

(i) a machine-transformation of transaction data that were entered so far by the user via said electronic device,

(ii) a machine-transformation of user-specific biometric characteristics that were extracted from analysis of said live video feed of the user,

(iii) a machine-transformation of video content of one or more video frames captured in said live video feed of the user.

12 . The method of claim 1 ,

wherein said encoded on-screen visual transformation comprises both:

(I) a machine-transformation of transaction data that were entered so far by the user via said electronic device,

and also,

(II) at least one of:

(i) a machine-transformation of user-specific biometric characteristics that were extracted from analysis of said live video feed of the user,

(ii) a machine-transformation of an image of an identification card that the user holds towards the user-facing camera of the electronic device,

(iii) a machine-transformation of video content of one or more video frames captured in said live video feed of the user.

13 . The method of claim 1 ,

wherein said encoded on-screen visual transformation comprises both:

(I) a machine-transformation of transaction data that were entered so far by the user via said electronic device,

and also,

(II) outputs generated by one or more layers of a Machine Learning model that is pre-trained to receive as input video-frames of users and to generate as output data corresponding to user-specific characteristics.

14 . The method of claim 1 ,

wherein said encoded on-screen visual transformation includes a blended visual composite derived concurrently from (i) the user's face image, and (ii) an image of an identification document that the user holds towards the front-facing camera, and (ii) transactional data that the user has entered so far.

15 . The method of claim 1 ,

wherein said encoded on-screen visual transformation is computed on the full frame of the captured video image of the user prior to any cropping, segmentation, or feature extraction.

16 . The method of claim 1 ,

wherein said encoded on-screen visual transformation is generated by operating on one or more intermediate feature layers of a neural-network biometric model representing user-specific facial embeddings or texture patterns.

17 . The method of claim 1 ,

wherein the electronic device generates the encoded on-screen visual transformation using latent-space vectors that are produced by a Machine Learning model that is pre-trained to extract individualized biometric or behavioral traits of said user.

18 . The method of claim 1 ,

wherein said encoded on-screen visual transformation further incorporates or represents therein a pseudo-random code received at the electronic device from the remote server during execution of said electronic transaction.

19 . The method of claim 1 ,

wherein said encoded on-screen visual transformation further embeds and encodes therein: a nonce data-item or challenge token, that was sent from the server to the end-user device and that is valid only for a duration of an active ongoing verification session.

20 . The method of claim 1 ,

wherein said encoded on-screen visual transformation further incorporates therein a secret data-item that was previously stored in a secure storage of the end-user electronic device and that was cryptographically linked to a corresponding record maintained by said remote server.

21 . The method of claim 1 ,

wherein said encoded on-screen visual transformation further incorporates therein a secret data-item, that was previously stored in a secure storage of the end-user electronic device; wherein said secret data-item is also known to the remote server.

22 . The method of claim 1 ,

wherein said encoded on-screen visual transformation is displayed in a visual form that is not comprehendible by a human observer, and includes data embedded within a group of pixels in rendered image frames that are displayed on the screen of the electronic device of the user.

23 . The method of claim 1 ,

wherein said encoded on-screen visual transformation is visually obfuscated within the live video content.

24 . The method of claim 1 ,

wherein said encoded on-screen visual transformation is visually encoded within the live video content by spatially interlacing encoded pixel regions across multiple sequential frames of the shared video feed.

25 . The method of claim 1 ,

wherein said encoded on-screen visual transformation is displayed as a semi-transparent element blended into a graphical user interface for entry of transaction data, while also remaining machine-recoverable through algorithmic decoding of pixel intensity variations.

26 . The method of claim 1 ,

wherein said encoded on-screen visual transformation is displayed as a visible on-screen element selected from: barcode, QR code, a group of color-coded pixels.

27 . The method of claim 1 ,

wherein said encoded on-screen visual transformation is displayed as a visible on-screen element which is a dynamically-changing shape-shifting group of pixels that are rendered over a region of an interface for transaction data entry, while said user is entering transaction data trough said interface, and while the electronic device performs continuous screen-sharing of the screen of the electronic device towards the remote server.

28 . The method of claim 1 ,

wherein said encoded on-screen visual transformation comprises a non-static, non-fixed, dynamically changing group of pixels that form an animated abstract pattern whose parameters vary according to session-specific transactional data and according to user-specific characteristics.

29 . The method of claim 1 ,

wherein said encoded on-screen visual transformation visually resembles the user's captured image but includes controlled distortions or modifications that encode transaction verification data.

30 . The method of claim 1 ,

wherein said encoded on-screen visual transformation is rendered as an animated and non-static and dynamically-changing group-of-pixels that are presented as a background layer positioned behind a graphical user interface for entering of transaction data.

31 . The method of claim 1 ,

wherein said encoded on-screen visual transformation is rendered as an animated and non-static and dynamically-changing group-of-pixels that are presented near, and not behind, a graphical user interface for entering of transaction data.

32 . The method of claim 1 ,

wherein said encoded on-screen visual transformation is rendered as an animated and non-static and dynamically-changing group-of-pixels that are presented near or behind a graphical user interface for entering of transaction data; wherein the screen of the electronic device of the user shows said encoded on-screen visual transformation, and does not show a live feed of the selfie video, to preserve privacy of the user while also providing user-authentication data and transaction-verification data through said encoded on-screen visual transformation that is screen-shared with the trusted remote server.

33 . The method of claim 1 ,

wherein the encoded on-screen visual transformation is generated by a rendering engine that transforms intermediate feature maps into faceless animated on-screen blobs of pixels,

wherein size and/or shape and/or color intensity of said faceless animated on-screen blobs of pixels is dynamically modified in synchronization with at least one of: (i) user-detected liveness cues, (ii) transaction data entered so far by the user, (iii) user-specific characteristics extracted from video-frames by one or more layers of a Deep Learning neural network.

34 . A system comprising:

one or more hardware processors, configured to execute code;

which are operably associated with one or more memory units that are configured to store data;

wherein the one or more hardware processors are configured to perform a process comprising:

while a user interacts with an electronic device to enter transaction data for an electronic transaction intended to be performed online via a remote server,

(a) activating a user-facing camera of the electronic device,

and capturing a live video feed of the user while the user enters transaction data;

(b) generating locally on said electronic device an encoded on-screen visual transformation that is a machine-transformation of at least one of:

(b1) a machine-transformation of video content of one or more video frames captured in said live video feed of the user,

(b2) a machine-transformation of transaction data that were entered so far by the user via said electronic device,

(b3) a machine-transformation of user-specific biometric characteristics that were extracted from analysis of said live video feed of the user,

(b4) a machine-transformation of an image of an identification card that the user holds towards the user-facing camera of the electronic device;

(c) displaying on a screen of said electronic device: (c1) a graphical user interface having one or more fillable fields for entering transaction data, and also (c2) a dynamically-changing non-static group of pixels that depict said encoded on-screen visual transformation;

(d) continuously sharing the screen of said electronic device with a trusted remote server, which is configured to perform server-side analysis of content that the trusted remote server receives via Screen Sharing from said electronic device;

wherein the server-side analysis comprises:

(d1) decoding the dynamically-changing non-static group of pixels that depict said encoded on-screen visual transformation, as received at the trusted remote server via Screen Sharing from the electronic device of the user; wherein said decoding at the trusted remote server yields decoded information that comprises at least one of: decoded transaction data, decoded user data;

(d2) comparing the decoded data at the trusted remote server, against at least one of: (i) transaction data that the trusted remote server received from the electronic device via a communication channel other than Screen Sharing, (ii) user data that the trusted remote server received from the electronic device via a communication channel other than Screen Sharing;

(d3) based on said comparing, determining at the trusted remote server whether to block or approve a user-submitted transaction that corresponds to said transaction data.