I am currently using this example from flandmark using JavaCV. Refer to: https://github.com/bytedeco/javacv-examples/blob/master/flandmark-demo/src/main/java/flandmark/Example1.java. I managed to run the code but is there a way that I could display the coordinated of each of the key point from the face (eye, mouth, nose)? There are around 7/8 key points that could be detected by the flandmark library. Thanks.
You find landmarks in function detectFaceInImage
. At the bottom of the same function you see how to draw the detected landmarks on the face:
flandmark_detect(input, bbox, model, landmarks);
// display landmarks
cvRectangle(orig, cvPoint(bbox[0], bbox[1]), cvPoint(bbox[2], bbox[3]), CV_RGB(255, 0, 0));
cvRectangle(orig,
cvPoint((int) model.bb().get(0), (int) model.bb().get(1)),
cvPoint((int) model.bb().get(2), (int) model.bb().get(3)), CV_RGB(0, 0, 255));
cvCircle(orig,
cvPoint((int) landmarks[0], (int) landmarks[1]), 3, CV_RGB(0, 0, 255), CV_FILLED, 8, 0);
for (int i = 2; i < 2 * model.data().options().M(); i += 2) {
cvCircle(orig, cvPoint((int) (landmarks[i]), (int) (landmarks[i + 1])), 3, CV_RGB(255, 0, 0), CV_FILLED, 8, 0);
}
bbox
is a int[]
containing the coordinates of the face bounding box as:
int[0] -> top left x
int[1] -> top left y
int[2] -> bottom right x
int[3] -> bottom right y
landmarks
is a double[]
, containing the coordinates of the points of the landmark in sequence:
index: 0 1 2 3 ...
coord: s0x s0y s1x s1y ...
with reference to the image from the flandmark homepage
The total number of landmarks is stored in model.data().options().M()
. Now you have all the basics to print the coordinates of the landmarks:
for (int i = 0; i < 2 * model.data().options().M(); i += 2) {
System.out.println("S" + (i/2) + ": (" + (int)(landmarks[i]) + ", " + (int)(landmarks[i+1]) + ")");
}