Methods for image processing of face regions and electronic devices using the same

1. A face-slimming method applied to a series of images captured by an electronic device, comprising:

receiving a current image;

detecting at least a face landmark corresponding to a face region of the current image;

comparing the face landmark of the current image with a reference landmark;

adjusting a distortion mask according to the comparison result; and

adjusting a size of the face region within the current image according to the distortion mask, wherein the distortion mask corresponds to a position of the face landmark,

wherein the comparing of the face landmark comprises:

determining whether the position of the face landmark falls within a stable window; and

increasing an unstable count number in response to the position of the face landmark falling outside of the stable window, where the stable window is defined according to the position of the reference landmark,

wherein the comparing of the face landmark further comprises:

determining whether the unstable count number exceeds a predetermined threshold; and

updating the reference landmark with the face landmark of the current image in response to the unstable count number exceeding the predetermined threshold.

2. The face-slimming method of claim 1 , wherein the face landmark comprises a combination of a philtrum parameter, a cheek parameter, a chin parameter, and a face angle parameter.

3. The face-slimming method of claim 1 , wherein the adjustment of the distortion mask further comprises adjusting positions of the face landmark within the distortion mask.

4. The face-slimming method of claim 1 , wherein the receiving of the current image further comprises:

initiating a real-time video communication application; and

capturing the series of images including the current image by a camera module of the electronic device.

5. An electronic device, comprising:

a camera module, configured to capture a series of images, the series of images comprising a face of a user;

an image processing module, configured to receive the series of images and, for each incoming current image, determine at least one face landmark of a face region within the current image, compare the face landmark with a reference landmark and adjust a size of the face region according to the comparison result; and

an image output module, configured to transmit the series of the adjusted images,

wherein the image processing module comprises:

a spatial condition determination module, configured to determine whether the face landmark of the current image falls within a stable window, and increase an unstable count number in response to the face landmark falling outside of the stable window, the stable window being defined according to the reference landmark; and

a temporal condition determination module, configured to determine whether the unstable count number exceeds a predetermined threshold,

wherein the spatial condition determination unit is further configured to update the reference landmark in response to the unstable count number exceeding the predetermined threshold.

6. The electronic device of claim 5 , wherein the image processing module comprises:

a face detection module, configured to detect the face region within the series of images; and

a face landmark detection module, configured to determine a position of the face landmark of the face region.

7. The electronic device of claim 5 , wherein the image processing module comprises:

a distortion mask generation module, configured to adjust a distortion mask corresponding to the face region according to the comparison result; and

a pixel registration module, configured to adjust the size of the face region in the current image according to the distortion mask;

wherein the distortion mask corresponds to positions of the face landmark.

8. The electronic device of claim 5 , wherein the face landmark comprises a combination of a philtrum parameter, a cheek parameter, a chin parameter and a face angle parameter.

9. The electronic device of claim 5 , wherein the image output module is further configured encode the series of images into a video stream to be transmitted over a wireless network.

10. An image processing method for use in an electronic device, comprising:

capturing a series of images;

detecting a face region of a user from each of the images;

defining a distortion mask corresponding to the face region;

adjusting a size of the face region of each image within the distortion mask; and

encoding the adjusted images into a video stream,

wherein the step of defining the distortion mask further comprises:

detecting a face landmark for each image;

determining whether the detected face landmarks have successively moved beyond a tolerable range from a reference landmark by a predetermined number of images;

in response to the detected face landmark moved beyond the tolerable range, defining the distortion mask according to the detected face landmark; and

in response to the detected face landmark not moved beyond the tolerable range, defining the distortion mask according to the reference landmark,

wherein the determining step further comprises:

for each image, increasing an unstable count number in response to the detected face landmark falls outside of a stable window around the reference landmark; and

determining that the detected face landmarks have successively moved beyond the tolerable range from the reference landmark by the predetermined number of images and updating the reference landmark in response to the unstable count number exceeds a predetermined threshold.

11. The image processing method of claim 10 , wherein the distortion mask covers a face area below a philtrum position and a left cheek position and a right cheek position.

12. The image processing method of claim 10 , wherein the adjustment of the face region for each image further comprises:

determining a target pixel for each pixel within the distortion mask, wherein the target pixel is with a longer distance from the center of the distortion mask than that with the pixel; and

adjusting a color value of the pixel according to a color value of the target pixel.

13. The image processing method of claim 10 , wherein the adjusted image has a smaller size of the face region than the original image.