DefinePK

Abstract

This study presents a new high-capacity spatial steganographic scheme for 3D point cloud models using a Self-Similarity Position Matching (SSPM) procedure. The new scheme partitions the 3D point cloud model to patches, clusters patches into similarity patch chains using self-similarity measures and generates the codebook. The representative patches and similar patches are then taken from the codebook for every similar patch chain as the reference patches and the message patches. Finally, every message point in the similar message patches which has the point-to-point correspondence with a certain reference point in the reference patch can embed at least four bits using the proposed SSPM, which embeds information by shifting the message point from current point to the corresponding embedding position that is computed over virtual sphere with the reference point as the center. Experimental results show that the proposed technique is secure, has high capacity and low distortion and is robust against uniform affine transformations such as transformation, rotation, scaling. In addition, a concise shape description and a similarity measures are devoted to improving performance for forming codebook and constructing point traversal. The technique can be considered as a side-match steganography and has proven to be a feasible alternative to other steganographic schemes for 3D point cloud model.