A. Yu. Aksenov - Junior Research Scientist, Laboratory of Research Automation, St. Petersburg Institute for Informatics and Automation of the Russian Academy of Sciences (SPIIRAS). E-mail: a_aksenov@mail.iias.spb.su, sial.iias.spb.su
V. V. Alexandrova - Ph.D. (Ped.), Research Scientist, Laboratory of Research Automation St. Petersburg Institute for Informatics and Automation of the Russian Academy of Sciences (SPIIRAS). E-mail: alexandr@iias.spb.su
A. A. Zaytseva - Ph.D. (Eng.), Senior Research Scientist, Laboratory of Research Automation, St. Petersburg Institute for Informatics and Automation of the Russian Academy of Sciences (SPIIRAS). E-mail: cher@iias.spb.su

3D-data can be obtained in several ways: modeling, scanning, application of tomography 3D-recovery from several 2D images. Depending on data obtaining method the resulting bit volume of data will greatly differs. For efficient storage of 3D-objects the representation methods based on object class are needed. The paper proposes the method of point cloud compression based on space subdivision by 256×256×256 cells and it-s further bit stream representation based on space filling curve scanning and secondary compression. The point cloud obtained through 3D-scanning consists of compact regions of space which are corresponding to surfaces of scanned object, therefore the implementation of Gilbert-s space filling curve for 1D bit stream representation has great effect on secondary compression. The proposed method can be applied in 3D-scanning, processing and printing software, efficient storage of 3D-objects spatial data, systems of cognitive programming, reducing cost of data transmission.