Image optimization during facial recognition
Described is a technique for optimization an image for facial detection. More specifically, described is a process of predicting the location of a face within an image and adjusting image settings based on at least a portion of the predicted location of the face. An image may be adjusted based on the characteristics of a metering region, which may be selected prior to performing facial detection. For example, the metering region may be a specified shape with dimensions equal to a certain percentage of the input image and placed at a specified location. The result of using such a metering region is that the image adjustments may be based on a portion of the face, and therefore, may be optimized for facial detection.
1. A method comprising:
receiving, by a computing device, a live image captured by a camera, wherein the live image continuously captures an image within a field of view of the camera;
receiving, by the computing device, an indication to perform facial recognition on a frame of the live image;
selecting, by the computing device, in response to the indication, a metering region within a portion of the frame prior to attempting to locate a position of a face within any portion of the frame, wherein the selected metering region is smaller in size than a first metering region, the first metering region configured as part of a default capture setting of the camera;
adjusting, by the computing device, the frame based on the selected metering region; and
detecting, by the computing device, the face within the adjusted frame.
2. The method of claim 1 , wherein the selected metering region is based on an expected location of the face.
3. The method of claim 1 , wherein the selected metering region is static and determined prior to capturing the live image.
4. The method of claim 1 , wherein adjusting the frame includes adjusting an exposure setting.
5. The method of claim 1 , wherein adjusting the frame includes adjusting the capture setting of the camera, the capture setting including at least one of an automatic gain control (AGC), color balance, focus, contrast, saturation, brightness, gamma, hue, and sharpness.
6. The method of claim 1 , further comprising displaying a preview of the captured live image and wherein the preview includes a face positioning guide and the selected metering region is based on a location of the positioning guide.
7. The method of claim 1 , further comprising;
displaying a preview of the captured live image; and
receiving a user input selecting a portion of the preview and wherein the selected metering region is based on the received user input.
8. The method of claim 1 , further comprising identifying a low contrast region of the live image and wherein the selected metering region is within the low contrast region.
9. The method of claim 1 , wherein the indication to perform facial recognition includes a request to authenticate a face of a user.
10. The method of claim 1 , further comprising:
comparing facial features of the face to features in a stored reference template; and
determining, based on the comparison, a recognition confidence score.
11. A method comprising:
receiving, by a computing device, an image captured by a camera;
receiving, by the computing device, an indication to perform facial recognition on the image;
selecting, by the computing device, in response to the indication, a metering region within a portion of the captured image based on a set of coordinates, wherein the metering region is selected prior to locating a position of a face within the image, wherein the selected metering region is smaller in size than a first metering region, the first metering region configured as part of a default capture setting of the camera;
adjusting, by the computing device, the image based on the selected metering region; and
detecting, by the computing device, the face within the adjusted image.
12. The method of claim 11 , wherein the image is a live image.
13. The method of claim 11 , wherein the selected metering region is centered within the image.
14. The method of claim 11 , wherein the set of coordinates is determined prior to capturing the image.
15. The method of claim 11 , wherein the selected metering region size is greater than or equal to 5% of the image size and less than or equal to 50% of the image size.
16. The method of claim 11 , wherein the selected metering region includes at least two quadrants of the image.
17. The method of claim 11 , further comprising:
comparing facial features of the face to features in a stored reference template; and
determining, based on the comparison, a recognition confidence score.
18. The method of claim 17 , further comprising:
making a determination that the recognition confidence score is above a threshold value; and
in response to determination, identifying a user associated with the facial features.
19. A method comprising:
receiving, by a computing device, a live image captured by a camera, wherein the live image continuously captures an image within a field of view of the camera;
receiving, by the computing device, an indication to perform facial recognition on a frame of the live image;
during a facial detection step of the facial recognition, selecting, by the computing device, in response to the indication, a metering region within a portion of the frame prior to attempting to locate a position of a face within any portion of the frame, wherein the selected metering region is smaller in size than a first metering region, the first metering region configured as part of a default capture setting of the camera;
adjusting, by the computing device, the frame based on the selected metering region; and
detecting, by the computing device, the face within the adjusted frame.