False face representation identification
Systems and methods for identifying a false representation of a human face are provided. In one example, a method for identifying a false representation of a human face includes receiving a plurality of different data streams captured by a respective plurality of sensors of differing sensor types sensing a candidate face. In a cascading plurality of stages, one or more of the different data streams are analyzed, wherein each of the stages comprises a different analysis. In one of the cascading plurality of stages, the method determines that one or more of the different data streams corresponds to a false representation of the human face. Based on determining that one or more of the different data streams corresponds to a false representation of a human face, an indication of the false representation is outputted.
1. A method for identifying a false representation of a human face, the method comprising:
receiving one or more data streams captured by one or more sensors sensing a candidate face;
sequentially detecting various spoofing attempts in a cascading plurality of stages to determine that one or more of the data streams corresponds to the false representation of the human face, wherein the cascading plurality of stages are arranged in a predetermined order of decreasing commonality of the spoofing attempts beginning with more common spoofing attempts and ending with less common spoofing attempts; and
based at least in part on determining that the one or more data streams corresponds to the false representation of the human face, outputting an indication of the false representation of the human face.
2. The method of claim 1 , wherein the one or more data streams comprise an infrared data stream, and the stages comprise determining whether an infrared candidate face pattern of the candidate face matches a predetermined infrared human face pattern.
3. The method of claim 1 , wherein the stages comprise determining whether a candidate face size of the candidate face is within a predetermined range of human face sizes.
4. The method of claim 3 , wherein determining whether the candidate face size of the candidate face is within a predetermined range of human face sizes further comprises determining whether an interpupillary distance of the candidate face falls within a predetermined range of an average human interpupillary distance.
5. The method of claim 1 , wherein the stages comprise determining whether a plurality of candidate face depth points lies on a single flat plane or a curving plane.
6. The method of claim 1 , wherein the stages comprise analyzing a plurality of image frames of the candidate face for changes in color that indicate blood flow in the candidate face.
7. The method of claim 1 , wherein the stages comprise:
extracting a candidate face texture feature from one the data streams; and
determining whether the candidate face texture feature corresponds to a human face texture feature.
8. The method of claim 1 , wherein the stages comprise determining whether a high frequency component loss in one or more of the data streams exceeds a predetermined loss threshold.
9. A computing device for identifying a false representation of a human face, the computing device comprising:
one or more sensors configured to sense a candidate face;
a false representation detection program executed by a processor of the computing device, the false representation detection program configured to:
receive one or more data streams captured by the one or more sensors;
sequentially detect various spoofing attempts in a cascading plurality of stages to determine that one or more of the data streams corresponds to the false representation of the human face, wherein the cascading plurality of stages are arranged in a predetermined order of escalating difficulty of detecting the spoofing attempts; and
based at least in part on determining that one or more of the data streams corresponds to the false representation of the human face, output an indication of the false representation of the human face.
10. The computing device of claim 9 , wherein the stages are arranged in order of escalating complexity associated with detecting the spoofing attempts.
11. The computing device of claim 9 , wherein a level of difficulty of detecting one of the spoofing attempts corresponds to one or more of a frequency of successful identifications of the false representation of the human face and a frequency of unsuccessful identifications of the false representation of the human face.
12. The computing device of claim 9 , wherein the one or more data streams comprise an infrared data stream, and the stages comprise determining whether an infrared candidate face pattern of the candidate face matches a predetermined infrared human face pattern.
13. The computing device of claim 9 , wherein the stages comprise determining whether an interpupillary distance of the candidate face falls within a predetermined range of an average human interpupillary distance.
14. The computing device of claim 9 , wherein the stages comprise determining whether a plurality of candidate face depth points lies on a single flat plane or a curving plane.
15. The computing device of claim 9 , wherein the stages comprise analyzing a plurality of image frames of the candidate face for changes in color that indicate blood flow in the candidate face.
16. The computing device of claim 9 , wherein the stages comprise:
extracting a candidate face texture feature from one the data streams; and
determining whether the candidate face texture feature corresponds to a human face texture feature.
17. The computing device of claim 9 , wherein the stages comprise determining whether a high frequency component loss in one or more of the data streams exceeds a predetermined loss threshold.
18. A method for identifying a false representation of a human face, the method comprising:
receiving one or more data streams captured by the one or more sensors;
sequentially detecting various spoofing attempts in a cascading plurality of stages to determine that one or more of the data streams corresponds to the false representation of the human face, wherein the cascading plurality of stages are arranged in a predetermined order of escalating difficulty of detecting the spoofing attempts; and
based at least in part on determining that one or more of the data streams corresponds to the false representation of the human face, outputting an indication of the false representation of the human face.
19. The method of claim 18 , wherein the stages are arranged in order of escalating complexity associated with detecting the spoofing attempts.
20. The method of claim 18 , wherein a level of difficulty of detecting one of the spoofing attempts corresponds to one or more of a frequency of successful identifications of the false representation of the human face and a frequency of unsuccessful identifications of the false representation of the human face.