Depth accuracy of the stereo camera

I am doing a project on calibrating stereo ZED camera and finding its accuracy and compare with the Manufacturer's accuracy of 1% at 1m depth accuracy. For this purpose , the formula to calculate the depth accuracy is $dz = (z^2 * de) / (f * b)$ but how do we calculate $z$ , $de$ and $f$. Is is taken from matlab stereo-callibration app which gives 'Stereoparameter' ?

$dz$ is the depth error in meters, $z$ is the depth in meters, $de$ is the disparity error in pixels, $f$ is the focal length of the camera in pixels and $b$ is the camera baseline in meters.

• Can you name the parameters?
– Ben
Dec 16 '16 at 2:53
• dz is the depth error in meters, z is the depth in meters, de is the disparity error in pixels, f is the focal length of the camera in pixels and b is the camera baseline in meters. Dec 16 '16 at 3:48
• You have to "estimate" your disparity error somehow (depending you the stereo matching algorithm used?) and you should be set. f and b are camera constants and z is the depth in which you make the measurement. Dec 16 '16 at 23:21
• also see this answer: robotics.stackexchange.com/a/9714/1021 Dec 16 '16 at 23:22
• @Kozuch Thanks for replying. I have just callibrated 11 set of stereo images in matlab stereo calibration app and have got intrinsic and extrinsic parameters . Is it possible to get the disparity error from this parameter ?. As I am using the images from ZED stereo camera the b is 120mm and FOV is known 110degree , so how do I get the focal length in this case ? Dec 17 '16 at 1:20