Hybrid deep learning method for gender classification

1. A computer-implemented method for gender classification by applying a hybrid of feature learning and feature engineering to face images, comprising:

conducting feature learning on a face image by one or more computer processors, comprising:

feeding the face image into a first convolution neural network to obtain a first decision;

conducting feature engineering on a face image by the one or more computer processors, comprising:

automatically detecting facial landmarks in the face image;

describing each of the facial landmarks by a set of facial features;

transforming the facial features into a two-dimensional matrix; and

feeding the two-dimensional matrix into a second convolution neural network to obtain a second decision;

computing a hybrid decision based on the first decision and the second decision comprising:

multiplying the first decision by a first weight to produce a first weighted decision; and

multiplying the second decision by a second weight to produce a second weighted decision; and

updating the first weight and the second weight in the hybrid decision by backpropagation,

wherein the hybrid decision is computed based on the first weighted decision and the second weighted decision; and

classifying gender of the face image in accordance with the hybrid decision.

2. The computer-implemented method of claim 1 , wherein the facial landmarks describe at least a portion of a chin and an edge of a face.

3. The computer-implemented method of claim 1 , wherein the hybrid decision is an average, a sum, or a root-mean square function of the first weighted decision and the second weighted decision.

4. The computer-implemented method of claim 1 , wherein each of the facial features is described by a pair of cartesian coordinates.

5. The computer-implemented method of claim 1 , further comprising:

automatically detecting a face in a digital image by the one or more computer processors;

extracting a face portion surrounding the face from the digital image by the one or more computer processors; and

normalizing the face portion to obtain the face image.

6. The computer-implemented method of claim 1 , wherein the facial landmarks respectively describe at least a portion of an eye, an eyebrow, a mouth, a chin, an edge of a face, or a nose in the face image.

7. The computer-implemented method of claim 1 , wherein the facial landmarks include a plurality of groups each of which describes at least a portion of a facial feature.

8. The computer-implemented method of claim 7 , wherein the facial landmarks include an eye, an eyebrow, a mouth, a chin, an edge of a face, or a nose in the face image.

9. The computer-implemented method of claim 1 , wherein each of the facial features is described by a pair of coordinates.

10. The computer-implemented method of claim 1 , wherein the step of transforming the facial features into a two-dimensional matrix comprises:

expressing each of the facial features by a pair of coordinates; and

forming a first matrix using the facial features and their respective coordinates.

11. The computer-implemented method of claim 10 , further comprising:

concatenating copied of the first matrix one or more times to produce the two-dimensional matrix.

12. The computer-implemented method of claim 1 , wherein the two-dimensional matrix is a square matrix.