Data returned on "PictureDetected" event?

Question

I am using the visionrecognition module to detect learned object through images (using the following example: http://doc.aldebaran.com/2-4/dev/python/examples/vision/vision_recognition.html) and I want to localize the position of the object whenever a known object is detected, the value variable in the callback (from the function:def on_picture_detected(self, value)) returns data such as :

[[6887L, 546148L], [[None, 11L, 0.05116279050707817, [[0.3418557941913605, -0.07902733981609344], [-0.2096691131591797, -0.1323433220386505], [-0.30789878964424133, 0.2983075678348541], [0.3219611942768097, 0.42371243238449097], [0.3418557941913605, -0.07902733981609344]]]]]

I want to understand the meaning of these values (since this is not well documented) and I want to know whether it is possible to locate the detect object in a cartesian reference (x,y,z) or any other means? Is there a way to capture the distance between the robot and the detected picture?