Revision history [back]

Basically you did not listen to me in previous posts. In order to define depth from a 2D image, you need to use the idea of stereo vision which gives you the projection matrix. There is no simple formula to deform your rectangle into a 3D location based on a single bounding box...

Go read these documentations:

• http://blog.martinperis.com/2011/08/stereo-vision-why-camera-calibration-is.html
• http://docs.opencv.org/modules/calib3d/doc/calib3d.html
• http://docs.opencv.org/doc/tutorials/calib3d/camera_calibration_square_chess/camera_calibration_square_chess.html#cameracalibrationsquarechessboardtutorial