Analyzing facial recognition data and social network data for user authentication
Tools, strategies, and techniques are provided for evaluating the identities of different entities to protect business enterprises, consumers, and other entities from fraud by combining biometric activity data with facial recognition data for end users. Risks associated with various entities can be analyzed and assessed based on a combination of user liveliness check data, facial image data, social network data, and/or professional network data, among other data sources. In various embodiments, the risk assessment may include calculating an authorization score or authenticity score based on different portions or combinations of the collected and processed data.
1. A computer-implemented method, comprising:
performing, via a processor, a liveliness check by:
detecting that a head of a user is aligned with a camera operably coupled to the processor,
causing display of a phrase to the user,
causing output of instructions to the user to read the phrase aloud after a countdown period,
causing recording of a video of the user reading the phrase, to produce a recorded video, and
causing display of a progress bar during the recording of the video of the user reading the phrase, the progress bar indicating an amount of progress in capturing a video loop including facial data changes;
calculating, via a software application running on the processor, a first score for the user based on the liveliness check;
determining that the first score is insufficient for authentication of the user; and
combining the first score with a second score to form a combined score, in response to determining that the first score is insufficient for authentication of the user, the second score being calculated based on social network connections of the user.
2. The method of claim 1 , wherein the liveliness check includes detecting movement of at least one lip of the user.
3. The method of claim 1 , wherein the liveliness check includes detecting at least one of: an eye blink, an eye movement, a head movement, a lip movement, a hand movement, or an arm movement, based on the recorded video.
4. The method of claim 2 , wherein the movement of at least one lip of the user occurs during a predefined time period of the recorded video of the user.
5. The method of claim 1 , further comprising verifying a transaction in response to the combined score exceeding a predefined value.
6. A system, comprising:
a processor; and
a memory operably coupled to the processor and storing instructions executable by the processor to:
perform a liveliness check by:
detecting that a head of a user is aligned with a camera operably coupled to the processor,
causing display of a phrase to the user,
causing output of instructions to the user to read the phrase aloud after a countdown period,
causing recording of a video of the user reading the phrase, to produce a recorded video, and
causing display of a progress bar during the recording of the video of the user reading the phrase, the progress bar indicating an amount of progress in capturing a video loop including facial data changes;
calculate a first score for the user based on the liveliness check; and
combine the first score with a second score to form a combined score, in response to determining that the first score is insufficient for authentication of the user, the second score being calculated based on a plurality of social network connections of the user.
7. The system of claim 6 , wherein the instructions executable by the processor to perform the liveliness check include instructions to movement of at least one lip of the user.
8. The system of claim 6 , wherein the instructions executable by the processor to perform the liveliness check include instructions to detect at least one of: an eye blink, an eye movement, a head movement, a lip movement, a hand movement, or an arm movement.
9. The system of claim 7 , wherein the movement of the at least one lip of the user occurs within a predefined time period of the recorded video of the user.
10. A non-transitory computer-readable medium storing processor-executable instructions to:
perform a liveliness check by:
detecting that a head of a user is aligned with a camera operably coupled to the processor,
causing display of a phrase to the user,
causing output of instructions to the user to read the phrase aloud after a countdown period,
causing recording of a video of the user reading the phrase, to produce a recorded video, and
causing display of a progress bar during the recording of the video of the user reading the phrase, the progress bar indicating an amount of progress in capturing a video loop including facial data changes;
calculate a first score for the user based on the liveliness check; and
combine the first score with a second score to form a combined score when the first score is determined insufficient for authentication of the user, the second score being a normalized weighted score calculated based on social network connections of the user.
11. The computer-readable medium of claim 10 , wherein the processor-executable instructions to perform the liveliness check include instructions to detect movement of at least one lip of the user.
12. The computer-readable medium of claim 10 , wherein the processor-executable instructions to perform the liveliness check include instructions to detect at least one of: an eye blink, an eye movement, a head movement, a lip movement, a hand movement, or an arm movement.
13. The computer-readable medium of claim 11 , wherein the movement of the at least one lip of the user occurs within a predefined time frame of the recorded video of the user.
14. The computer-readable medium of claim 10 , further storing processor-executable instructions to verify a transaction in response to the combined score exceeding a predefined value.
15. The method of claim 1 , wherein the phrase is a randomly generated phrase.
16. The method of claim 1 , further comprising:
extracting a frame from the video of the user; and
generating a facial vector map based on the frame.