Wavelet-based high-capacity watermarking of 3-D irregular meshes

Abstract

International audienceDigital watermarking can be used as data hiding technique to interleave cover content with auxiliary information before transmitting and storing applications. While image and video watermarking has been widely studied, much less attention has been paid to its application in 3D mesh models. This is principally due to their intrinsic irregular sampling nature. This paper proposes a high-capacity watermarking scheme for the purpose of inserting meta-data into 3D triangle meshes. Our proposal can be applied to meshes with arbitrary topology by using irregular wavelet-based analysis. The watermark is embedded in an appropriate resolution level by quantizing the norms of wavelet coefficient vectors. To ensure robustness to similarity transformation, a robust synchronization (indexing) mechanism is performed on the 3D model after irregular wavelet analysis. Experimental results show that our watermarking framework is robust to common geometric attacks and can provide relatively high data embedding rate whereas keep a relative lower distortion

Steganalysis of meshes based on 3D wavelet multiresolution analysis

“…w l (i,j) and the edge vector, e l (i,j) , from the lower resolution mesh representation are considered as components of the proposed 3D wavelet feature vector used for steganalysis. The reason we have chosen these WCVs and their support edge vectors as steganalytic features is because these are changed when hiding information by the steganographical algorithms proposed in [20,24,25].…”

“…Under this framework, a given mesh is represented by a coarser mesh and a set of vectors, called Wavelet Coefficient Vectors (WCV), used for encoding the details. 3D wavelet analysis was used for various applications including filtering [16], mesh compression [17,18], subdivision [19], as well as information hiding [20,21,22,23,24,25]. A 3D wavelet-based watermarking algorithm was proposed by Kanai et al [20], which modifies the ratio between the norm of a WCV and the length of its support edge.…”

“…A hierarchical 3D watermarking methodology based on the wavelet transform was developed by Wang et al [24], which includes three different algorithms, each one enforcing one of the following requirements: robustness, high-capacity and fragility for authentication. Zaid et al [25] proposed a high-capacity 3D wavelet-based watermarking algorithm, by applying Quantization Index Modulation (QIM) [26] on the WCVs in order to embed information. More details about the 3D wavelet-based information hiding algorithms are given in Section 2.…”

A multi-secret sharing scheme for 3D solid objects

“…of three-dimensional data and objects. Unfortunately, there have proposed few algorithms dedicated to this purpose and the majority of them are related to watermarking and encryption techniques: Lee, Kwon, and Kwon (2014) introduced a multiple-watermarking scheme for copyright protection of mobile 3D contents with low resolution; Zaid, Hachani, and Puech (2014) proposed a high-capacity watermarking scheme for the purpose of inserting meta-data into 3D triangle meshes; an error detecting code based reversible fragile watermarking scheme for 3D models in spatial domain was proposed by Wang, Fan, Huang, and Li (2014), Muniraj, Kim, and Lee (2014) proposed a Fourier-based encryption method for volumetric 3D objects; Cho and Javidi (2013) presented a threedimensional photon counting double-random-phase encryption technique for 3D scenes; Lee and Cho (2013) introduced am optical encryption and information authentication of 3D objects considering wireless channel characteristics. Others works have been proposed earlier (Ben Hamza, 2009;Chou & Tseng, 2006;Kishk & Javidi, 2003;Lee & Kwon, 2011;Liu, Hu, Zhang, Zhu, & Li, 2011;Montañola Sales, Rondão Alface, & Macq, 2011;Wang, Lavoué, Denis, & Baskurt, 2007;Yeo & Yeung, 1999).…”