Sahab Matlab contest

Depth cameras give a 3D distance measurement for each image pixel. These sensors are increasingly popular in Engineering applications such as automation, manufacturing, structural surveys, robotics, computer vision, 3D markerless motion capture and home entertainment. The image on the left shows a Microsoft Kinect sensor (which includes a depth camera) pointing at the corner of a room. The orange rectangle gives a rough boundary of the depth image captured by the sensor. The depth image is provided as separate metric x,y,z values for each pixel in the image. These values are available from the text files named x.txt, y.txt and z.txt.
The goal of this question is to write a MATLAB program that will extract the left and right walls from the depth image and visually identify the door handle (which protrudes from the wall). The latter is extremely useful for any scenario where an automatic system must find objects that do not belong to a plane (detecting manufacturing defects, intelligent robotics, object tracking, 3D modelling etc)