I am new to openCV, I want to determine the distance between the android phones front camera and the face detected.
Is there any way to get the dimensions of the face?
Thanks.
Find out what is the average distance between human eyes, eyes and mouth and based on that and your camera resolution you can estimate the camera-face distance.
I don't think there is/you need a special formula for that. If, say, average eye distance is 5 cm, get something that is 5 cm long or print a pattern with two dots 5 cm apart. Then take a few snapshots from known distances (say, 50 cm apart) and measure the eye distance/dots displacement in pixels. Cover the entire range within which you will identify faces - this will not be too far, probably a few meters max. Put the data into a chart and you will see the relation of the detected eye distance in pixels to the distance of the observed face from the camera. Average face. Face that is looking straight into the camera. To get more robust results, use eye-mounth distance as well. For far away faces, when you cannot identify eyes, use the entire size of the head/face for rough results.
Hi Nishant,
You will be needing stereo images to find the distance.Calibrate your camera using OpenCV only. Get the intrinsic parameters which include the focal length and all. Create a disparity map using StereoBM or Stereo_SGBM, With the help of that depth map try to find the final distance.
As requested these links might help you:
http://docs.opencv.org/modules/calib3d/doc/camera_calibration_and_3d_reconstruction.html
http://wiki.ros.org/camera_calibration/Tutorials/StereoCalibration
http://blog.martinperis.com/2011/01/opencv-stereo-camera-calibration.html
http://abhishek4273.wordpress.com/2014/07/09/playing-with-stereo-images-and-depth-map/
http://abhishek4273.wordpress.com/2014/02/20/camera-calibration-using-opencv/
Asked: 2014-07-08 14:29:00 -0600
Seen: 4,546 times
Last updated: Jul 14 '14
