System, method, and device of authenticating a user based on selfie image or selfie video
System, method, and device of detecting identity of a user and authenticating a user; as well as detecting a possible attacker or impostor, and differentiating among users of an electronic device or of a computerized service. A mobile or portable electronic device is utilized to capture a self-taken image or video of a user, which is utilized as a user-authentication factor. The accelerometer and gyroscope or device-orientation sensor of the mobile device, sense and measure spatial and physical device properties during, before or after the submission of the self-taken image or video. Based on such spatial and physical device properties, in combination with computer-vision analysis of the content shown in the self-taken image or video, the system determines liveness of the user and freshness of the submitted self-taken image or video, and differentiates between a legitimate user and an attacker.
1. A method comprising:
(a) generating a user-authentication process for authenticating a user of an electronic device,
wherein the user-authentication process utilizes a self-taken visual-acquisition input of said user,
wherein the self-taken visual-acquisition input comprises one of: a self-taken image, a self-taken video;
(b) performing image recognition on content shown in said self-taken visual-acquisition input, and extracting a particular feature based on image recognition of said content;
(c) determining a characteristic of said electronic device, based on data sensed by one or more non-camera sensors of said electronic device during acquisition of said self-taken visual-acquisition input;
(d) detecting a mismatch between (I) said particular feature that was extracted in step (b) based on image recognition of the content of the self-taken visual-acquisition input, and (II) said characteristic of the electronic device that was determined in step (c) based on data sensed by said one or more non-camera sensors of said electronic device;
(e) based on said mismatch detected in step (d), denying an authentication request received from said electronic device
wherein step (d) comprises:
detecting that the content of the self-taken visual-acquisition input, indicates that the electronic device was moved during acquisition at a movement range of N degrees;
detecting that one or more non-camera sensors of the electronic device, indicate that the electronic device was moved during acquisition at a movement range of M degrees,
wherein the difference between M and N is at least P percent of N,
wherein P and M and N are pre-defined values;
based on the above detecting operations, determining said mismatch, and denying said authentication request.
2. The method of claim 1 ,
wherein step (d) comprises:
detecting that the content of the self-taken visual-acquisition input, indicates that the electronic device was held vertically during acquisition;
detecting that a spatial-orientation sensor of the electronic device, indicates that the electronic device was held non-vertically during acquisition;
based on the above detecting operations, determining said mismatch, and denying said authentication request.
3. The method of claim 1 ,
wherein step (d) comprises:
detecting that the content of the self-taken visual-acquisition input, indicates that the electronic device was held horizontally during acquisition;
detecting that a spatial-orientation sensor of the electronic device, indicates that the electronic device was held non-horizontally during acquisition;
based on the above detecting operations, determining said mismatch, and denying said authentication request.
4. The method of claim 1 ,
wherein step (d) comprises:
determining that there exists a contradiction between (I) spatial orientation of the electronic device as sensed by one or more non-camera spatial orientation sensors of the electronic device during acquisition, and (II) spatial orientation of the electronic device as determined from computer vision analysis of content shown in said self-taken visual-acquisition input;
based on said determining, denying said authentication request.
5. The method of claim 1 ,
wherein step (d) comprises:
determining that there exists a contradiction between (I) spatial orientation of the electronic device as sensed by one or more gyroscope sensors of the electronic device during acquisition, and (II) spatial orientation of the electronic device as determined from computer vision analysis of content shown in said self-taken visual-acquisition input;
based on said determining, denying said authentication request.
6. The method of claim 1 ,
wherein step (d) comprises:
determining a contradiction between
(A) output of an image recognition analysis applied to said self-taken visual-acquisition input, and
(B) data sensed by one or more of (B1) accelerometer of the electronic device, (B2) gyroscope of the electronic device, (B3) spatial-orientation sensor of said electronic device, (B4) Global Positioning System (GPS) sensor of the electronic device, (B5) location-determining sensor of the electronic device;
based on said contradiction, denying said authentication request.
7. The method of claim 1 ,
wherein the user authentication process, that is based on said self-taken visual-acquisition input, is configured to authorize or deny user access to at least one of: said electronic device, an electronic mail account, a social network account, an online retailer account, an online banking account, a vehicle, a physical venue.
8. A non-transitory storage medium having stored thereon instructions that, when executed by one or more hardware processors, cause said one or more hardware processors to perform a method comprising:
(a) generating a user-authentication process for authenticating a user of an electronic device,
wherein the user-authentication process utilizes a self-taken visual-acquisition input of said user,
wherein the self-taken visual-acquisition input comprises one of: a self-taken image, a self-taken video;
(b) performing image recognition on content shown in said self-taken visual-acquisition input, and extracting a particular feature based on image recognition of said content;
(c) determining a characteristic of said electronic device, based on data sensed by one or more non-camera sensors of said electronic device during acquisition of said self-taken visual-acquisition input;
(d) detecting a mismatch between (I) said particular feature that was extracted in step (b) based on image recognition of the content of the self-taken visual-acquisition input, and (II) said characteristic of the electronic device that was determined in step (c) based on data sensed by said one or more non-camera sensors of said electronic device;
(e) based on said mismatch detected in step (d), denying an authentication request received from said electronic device;
wherein step (d) comprises:
detecting that the content of the self-taken visual-acquisition input, indicates that the electronic device was moved during acquisition at a movement range of N degrees;
detecting that one or more non-camera sensors of the electronic device, indicate that the electronic device was moved during acquisition at a movement range of M degrees,
wherein the difference between M and N is at least P percent of N,
wherein P and M and N are pre-defined values;
based on the above detecting operations, determining said mismatch, and denying said authentication request.
9. A system comprising:
one or more processors, operably associated with one or more memory units,
wherein the one or more processors are configured:
(a) to generate a user-authentication process for authenticating a user of an electronic device,
wherein the user-authentication process utilizes a self-taken visual-acquisition input of said user,
wherein the self-taken visual-acquisition input comprises one of: a self-taken image, a self-taken video;
(b) to perform image recognition on content shown in said self-taken visual-acquisition input, and to extract a particular feature based on image recognition of said content;
(c) to determine a characteristic of said electronic device, based on data sensed by one or more non-camera sensors of said electronic device during acquisition of said self-taken visual-acquisition input;
(d) to detect a mismatch between (I) said particular feature that was extracted in step (b) based on image recognition of the content of the self-taken visual-acquisition input, and (II) said characteristic of the electronic device that was determined in step (c) based on data sensed by said one or more non-camera sensors of said electronic device;
(e) based on said mismatch, to deny an authentication request received from said electronic device;
wherein in (d) the one or more processors are configured:
to detect that the content of the self-taken visual-acquisition input, indicates that the electronic device was moved during acquisition at a movement range of N degrees;
to detect that one or more non-camera sensors of the electronic device, indicate that the electronic device was moved during acquisition at a movement range of M degrees,
wherein the difference between M and N is at least P percent of N,
wherein P and M and N are pre-defined values;
based on the above, to determine said mismatch, and to deny said authentication request.