Android Detection Preferences
Use Detection Preferences to enable and configure facial detection characteristics. Only the basic preferences are configurable by default; you can expose and configure the advanced preferences by clicking on the
icon in the upper right corner and selecting Show Advanced Preferences.
These are the default configurable preferences.
- Enable Face Detector: Enables face detection. Note that face detection must be enabled for face recognition to occur.
- Min Required Face Size: Defines the minimum required size for a face to be detected. Any face smaller than the height or width is ignored.
- Enable Liveness Detector: Enables RGB liveness detection.
- Activate only when recognized faces are present in the scene: When enabled, the Mobile Client only tests for RGB liveness detection for recognized faces.
- Detection Scheme: Specifies which RGB liveness model(s) should be used.
- Texture Unimodal: Only the Texture model will be used.
- Context Unimodal: Only the Context model will be used.
- Strict Multimodal: Both the Texture and Context models will be used. For a subject to pass the RGB liveness test, both of the results of the models must meet or exceed the Liveness detection threshold value. This is the default option.
- Normal Multimodal: Both the Texture and Context models will be used. For a subject to pass the RGB liveness test, the average of the results of the two models must meet or exceed the Liveness detection threshold value.
- Tolerant Multimodal: Both the Texture and Context models will be used. Subjects pass the RGB liveness test when the result of either model meets or exceeds the Liveness detection threshold value.
The advanced preferences can be exposed and configured by clicking on the
icon in the upper right corner and selecting Show Advanced Preferences.
The Enable Face Detector check box must be selected to enable face recognition.
- Enable Face Detector: Enables face detection. Note that face detection must be enabled for face recognition to occur.
- Max Vertical Resolution: Specifies the maximum supported vertical resolution.
- Min Searched Face Size: Defines the minimum face size that can be detected. A searched size of 80, for example, can still manage to detect faces as small as 60x60, but with lower certainty. Lowering this number enables SAFR to detect much smaller faces but also greatly increases CPU usage.
Note: This setting does not impact face recognition accuracy.
- Min Required Face Size: Defines the minimum required size for a face to be detected. Any face smaller than the height or width is ignored.
- Generate Recognizer Hint: Optimizes facial recognition. It should be turned on for most cases. If it is turned off, recognition accuracy may be reduced if detection is performed at very low resolutions.
- Detection Service: Specifies which face detection service will be used.
- Standard: The standard facial detection model that SAFR uses.
- High Sensitivity: A high sensitivity facial detection model which has a lower latency and whose performance doesn't degrade when multiple faces are being analyzed simultaneously. This model consumes many more GPU resources than the Standard model.
- Input Size: This setting is only available if Detection service is set to High Sensitivity. Input Size allows you to manage the trade-off between facial recognition accuracy vs. speed. There are 4 possible values:
- Normal: This is the standard against which the other 3 possible values are measured.
- Small: Decreased accuracy but increases speed.
- Extra Small: Greatly decreases accuracy but greatly increases speed.
- Large: Increases accuracy but decreases speed.
- Processors #: Specifies how many processors will be used for face detection. Selecting more processors results in more face detections per second, but it also consumes more system resources, which can slow everything down resulting in ultimately fewer recognition attempts per second.
- Detection Thresholds: Allows you to configure the detection thresholds.
- Enable Custom Thresholds: Allows you to customize the detection threshold.
- Initial Candidate Threshold: Initial face candidate threshold that is used during face detection.
- Middle Candidate Threshold: Middle face candidate threshold that is used during face detection.
- Final Candidate Threshold: Final face candidate threshold that is used during face detection.
- Detection Sensitivity Threshold: The lower this value is, the more lenient the facial detection service will be when attempting to recognize a face, which can result in additional false positives.
The Enable Liveness Detector check box must be selected to enable RGB liveness detection. RGB liveness detection is only available if you selected High Sensitivity for the Detection service setting above. For a full description of how RGB liveness works, please see the liveness detection topic.
- Enable Liveness Detector: Enables RGB liveness detection.
- Activate only when recognized faces are present in the scene: When enabled, the Mobile Client only tests for RGB liveness detection for recognized faces.
- Min Required Face Size: The minimum required height and width of a face, in number of pixels, for the Texture Model to be used.
- Minimum Required Face Context Size: The minimum required extra context around faces for the Context Model to be used.
- Minimum Required Center Pose Quality: The minimum face center pose quality for RGB liveness detection to be used.
- Minimum Required Sharpness Quality: The minimum face sharpness quality for RGB liveness detection to be used.
- Minimum Required Face Contrast Quality: The minimum face contrast quality for RGB liveness detection to be used.
- Minimum Preliminary Liveness Threshold: For multimodal detection schemes, this is the liveness threshold which the first evaluated model must exceed before SAFR will bother evaluating the second model. If this threshold is not met, SAFR immediately returns NOTLIVE_CONFIRMED for the subject.
- Liveness Detection Threshold: Specifies how difficult it is for a subject to be verified as LIVENESS_CONFIRMED.
- Fake Detection Threshold: Specifies how difficult it is for a subject to be verified as NOTLIVE_CONFIRMED.
- Detection Scheme: Specifies which RGB liveness model(s) should be used.
- Texture Unimodal: Only the Texture model will be used.
- Context Unimodal: Only the Context model will be used.
- Strict Multimodal: Both the Texture and Context models will be used. For a subject to pass the RGB liveness test, both of the results of the models must meet or exceed the Liveness detection threshold value. This is the default option.
- Normal Multimodal: Both the Texture and Context models will be used. For a subject to pass the RGB liveness test, the average of the results of the two models must meet or exceed the Liveness detection threshold value.
- Tolerant Multimodal: Both the Texture and Context models will be used. Subjects pass the RGB liveness test when the result of either model meets or exceeds the Liveness detection threshold value.
- Evaluate Liveness Over N Frames: The number of frames over which liveness should be evaluated.
- Evaluate Fake Over N Frames: The number of frames over which fakeness should be evaluated.
- Minimum Confirmations Required: The percentage of frames that must meet the liveness or fake threshold for the subject to be declared either LIVENESS_CONFIRMED or NOTLIVE_CONFIRMED.