Integral imaging (II) is a promising technique for sensing and visualizing of three dimensional (3D) images because it produces autostereoscopic images without special illumination requirements. As with any 3D imaging, in order to produce high quality 3D images, it is required to capture, record, transmit, process and display an enormous amount of optical data. Therefore, a central challenge that rises is to store and transmit efficiently the huge amount of information. This can be done by applying appropriate data compression techniques that remove efficiently the inherent redundancy within the captured data. In this work we survey previously developed II compression methods and, compare their performance. We present a new technique to cope with a common problem with some of the II compression methods, namely the choice of the compression depth to be applied in the various dimensions representing an II.