Publications

Book chapters

[4] Task-specific salience for object recognition Jérôme Revaud, Guillaume Lavoué, Yasuo Ariki and Atilla Baskurt , Chapter in Innovations in Intelligent Image Analysis , Series: Studies in Computational Intelligence, Vol. 339. Halina Kwasnicka, Lakhmi C. Jain (Editors), Springer-Verlag, ISBN: 978-3-642-17933-4, February 2011. (web) [3] Blind watermarking of three-dimensional meshes: Review, recent advances and future opportunities Kai Wang, Guillaume Lavoué, Florence Denis and Atilla Baskurt , Chapter in Advanced Techniques in Multimedia Watermarking: Image, Video and Audio Applications , A. Al-Haj (Editor), IGI Global, ISBN: 978-1615209033, April 2010. (web) [2] Basic background in 3D Object Processing Guillaume Lavoué, Chapter in 3D Object Processing: Compression, Indexing and Watermarking, Jean-Luc Dugelay, Atilla Baskurt, Mohamed Daoudi (Editors), John Wiley & Sons, ISBN: 978-0-470-06542-6, pp. 5-44, April 2008. (web) [1] 3D Compression Guillaume Lavoué, Florent Dupont and Atilla Baskurt, Chapter in 3D Object Processing: Compression, Indexing and Watermarking, Jean-Luc Dugelay, Atilla Baskurt, Mohamed Daoudi (Editors), John Wiley & Sons, ISBN: 978-0-470-06542-6, pp. 45-86,April 2008. (web)

International Journal Papers

• [19] Assessing Visual Quality of 3D Polygonal Models
  Abdullah Bulbul, Tolga Capin, Guillaume Lavoué , Marius Preda, IEEE Signal Processing Magazine, vol. 28, No. 6, pp. 80-90, 2011.(Abstract...) (Copyright IEEE)

This survey paper presents recent advances in evaluating and measuring the perceived visual quality of 3D polygonal models. The paper analyzes the general process of objective quality assessment metrics and subjective user evaluation methods and presents a taxonomy of existing solutions. Simple geometric error computed directly on the 3D models does not necessarily reflect the perceived visual quality; therefore, integrating perceptual issues for 3D quality assessment is of great significance. This paper discusses existing metrics, including perceptually based ones, computed either on 3D data or on 2D projections, and evaluates their performance for their correlation with existing subjective studies. (Hide...)

• [18] A Multiscale Metric for 3D Mesh Visual Quality Assessment  
  Guillaume Lavoué, Computer Graphics Forum (Proceedings of  Eurographics Symposium on Geometry Processing 2011), vol. 30, No. 5, pp. 1427-1437, 2011. (Abstract...) [Paper] [Code] (Copyright Blackwell)

  Many processing operations are nowadays applied on 3D meshes like compression, watermarking, remeshing and so forth; these processes are mostly driven and/or evaluated using simple distortion measures like the Hausdorff distance and the root mean square error, however these measures do not correlate with the human visual perception while the visual quality of the processed meshes is a crucial issue. In that context we introduce a full-reference 3D mesh quality metric; this metric can compare two meshes with arbitrary connectivity or sampling density and produces a score that predicts the distortion visibility between them; a visual distortion map is also created. Our metric outperforms its counterparts from the state of the art, in term of correlation with mean opinion scores coming from subjective experiments on three existing databases. Additionally, we present an application of this new metric to the improvement of rate-distortion evaluation of recent progressive compression algorithms. . (Hide...)

• [17] Learning boundary edges for 3D-mesh segmentation  
  Halim Benhabiles,  Guillaume Lavoué, Jean-Philippe Vandeborre and Mohamed Daoudi , Computer Graphics Forum , vol. 30, No. 8, pp. 2170-2182, 2011. (Abstract...) [Paper] [Material] (Copyright Blackwell)

  This paper presents a 3D-mesh segmentation algorithm based on a learning approach. A large database of manually segmented 3D-meshes is used to learn a boundary edge function. The function is learned using a classifier which automatically selects from a pool of geometric features the most relevant ones to detect candidate boundary edges. We propose a processing pipeline that produces smooth closed boundaries using this edge function. This pipeline successively selects a set of candidate boundary contours, closes them and optimizes them using a snake movement. Our algorithm was evaluated quantitatively using two different segmentation benchmarks and was shown to outperform most recent algorithms from the state-of-the-art. (Hide...)

• [16] Rate-distortion optimization for progressive compression of 3D mesh with color attributes  
  Ho Lee, Guillaume Lavoué and Florent Dupont, The Visual Computer, to appear, 2011. (Abstract...) [Paper] [Code] (Copyright Springer)

  We propose a new lossless progressive compression algorithm based on rate-distortion optimization for meshes with color attributes; the quantization precision of both the geometry and the color information is adapted to each intermediate mesh during the encoding/decoding process. This quantization precision can either be optimally determined with the use of a mesh distortion measure or be quasi-optimally decided based on an analysis of the mesh complexity in order to reduce the calculation time. Furthermore, we propose a new metric which estimates the geometry and color importance of each vertex during the simplification in order to preserve faithfully the feature elements. Experimental results show that our method outperforms the state-of-the-art algorithm for colored meshes and competes with the most efficient algorithms for non colored meshes. (Hide...)

• [15] Joint Reversible Watermarking and Progressive Compression of 3D Meshes  
  Ho Lee, Cagatay Dikici, Guillaume Lavoué and Florent Dupont, The Visual Computer (35 best papers from Computer Graphics International 2011), vol. 27, No. 6-8, pp. 781-792, 2011. (Abstract...) [Paper] [Code] (Copyright Springer)

  A new reversible 3D mesh watermarking scheme is proposed in conjunction with progressive compression. Progressive 3D mesh compression permits a progressive renement of the model from a coarse to a fine representation by using different levels of detail (LoDs). A reversible watermark is embedded into all reffinement levels such that (1) the reffinement levels are copyright protected, and (2) an authorized user is able to reconstruct the original 3D model after watermark extraction, hence reversible. The progressive compres- sion considers a connectivity-driven algorithm to choose the vertices that are to be rened for each LoD. The proposed watermarking algorithm modies the geom- etry information of these vertices based on histogram bin shifting technique. An authorized user can extract the watermark in each LoD and recover the original 3D mesh, while an unauthorized user which has access to the decompression algorithm can only reconstruct a distorted version of the 3D model. Experimental results show that the proposed method is robust to several at- tack scenarios while maintaining a good compression ratio. (Hide...)

• [14]  Robust and blind mesh watermarking based on volume moments
  Kai Wang, Guillaume Lavoué, Florence Denis and Atilla Baskurt, Computers & Graphics, Vol. 35, No. 1, pp. 1-19, 2011. (Abstract...) [Paper] (Copyright Elsevier)

  This paper presents a robust and blind watermarking algorithm for threedimensional (3D) meshes. The watermarking primitive is an intrinsic 3D shape descriptor: the analytic and continuous geometric volume moment. During watermark embedding, the input mesh is first normalized to a canonical and robust spatial pose by using its global volume moments. Then, the normalized mesh is decomposed into patches and the watermark is embedded through a modified scalar Costa quantization of the zero-order volume moments of some selected candidate patches. Experimental results and comparisons with the state of the art demonstrate the effectiveness of the proposed approach. . (Hide...)

• [13] A comparison of perceptually-based metrics for objective evaluation of geometry processing
  Guillaume Lavoué and Massimiliano Corsini, IEEE Transactions on Multimedia, Vol. 12, No. 7, pp. 636-649, 2010. (Abstract...) [Paper] [Datasets] [Code] (Copyright IEEE)

  Recent advances in 3D graphics technologies have led to an increasing use of processing techniques on 3D meshes, such as filtering, compression, watermarking, simplification, deformation and so forth. Since these processes may modify the visual appearance of the 3D objects, several metrics have been introduced so as to properly drive or evaluate them, from classic geometric ones such as Hausdorff distance, to more complex perceptually-based measures. This paper presents a survey on existing perceptually-based metrics for visual impairment of 3D objects and provides an extensive comparison between them. In particular, different scenarios which correspond to different perceptual and cognitive mechanisms are analyzed. The objective is twofold: (1) catching the behavior of existing measures so as to help Perception researchers for designing new 3D metrics and (2) providing a comparison between them so as to inform and help Computer Graphics researchers for choosing the accurate tool for the design and the evaluation of their mesh processing algorithms. (Hide...)

• [12] A comparative study of existing metrics for 3D-mesh segmentation evaluation
  Halim Benhabiles, Jean-Philippe Vandeborre, Guillaume Lavoué and Mohamed Daoudi, The Visual Computer, Vol. 26, No. 12, pp. 1451–1466, 2010. (Abstract...) [Paper] [Benchmark] (Copyright Springer)

  In this paper, we present an extensive experimental comparison of existing similarity metrics addressing the quality assessment problem of mesh segmentation. We introduce a new metric named the 3D Normalized Probabilistic Rand Index (3D-NPRI) which outperforms the others in terms of properties and discriminative power. This comparative study includes a subjective experiment with human observers and is based on a corpus of manually segmented models. This corpus is an improved version of our previous one~\cite{benhabiles09}. It is composed of a set of 3D-mesh models grouped in different classes associated with several manual ground-truth segmentations. Finally the 3D-NPRI is applied to evaluate six recent segmentation algorithms using our corpus and the Chen's et al.~\cite{chen09} corpus. (Hide...)

• [11] Semi-Sharp Subdivision Surface Fitting Based on Feature Lines Approximation
  Guillaume Lavoué and Florent Dupont, Computers & Graphics, Vol. 33, No. 2, pp. 151-161, 2009. (Abstract...) [video] [Paper] (Copyright Elsevier)

  This paper presents an algorithm for approximating arbitrary polygonal meshes with subdivision surfaces, with the objective of preserving the relevant features of the object while searching the coarsest possible control mesh. The main idea is to firstly extract the feature lines of the object, and secondly construct the subdivision surface over this network. Control points are created by approximating these lines while the connectivity is built with respect to the anisotropy of the object. Our algorithm reinforces the similarity between the subdivision surface and the original shape by affecting an integer sharpness degree to each control edge in order to accurately reproduce the different curvature radii of corresponding fillets and blends. (Hide...)

• [10] A Local Roughness Measure for 3D Meshes and its Application to Visual Masking
  Guillaume Lavoué, ACM Transactions on Applied Perception, Vol. 5, No. 4, Article 21, 2009. (Abstract...) [exe] [Paper] (Copyright ACM)

  3D models are subject to a wide variety of processing operations such as compression, simplification or watermarking, which may introduce some geometric artifacts on the shape. The main issue is to maximize the compression/simplification ratio or the watermark strength while minimizing these visual degradations. However few algorithms exploit the human visual system to hide these degradations, while perceptual attributes could be quite relevant for this task. Particularly, the masking effect defines the fact that one visual pattern can hide the visibility of another. In this context we introduce an algorithm for estimating the roughness of a 3D mesh, as a local measure of geometric noise on the surface. Indeed, a textured (or rough) region is able to hide geometric distortions much better than a smooth one. Our measure is based on curvature analysis on local windows of the mesh and is independent of the resolution/connectivity of the object. The accuracy and the robustness of our measure, together with its relevance regarding visual masking have been demonstrated through extensive comparisons with state-of-the-art and subjective experiment. Two applications are also presented, in which the roughness is used to lead (and improve) respectively compression and watermarking algorithms. (Hide...)

• [9] Improving Zernike Moments Comparison for Optimal Similarity and Rotation Angle Retrieval
  Jérôme Revaud, Guillaume Lavoué and Atilla Baskurt, IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 31, No. 4, pp. 627-636, 2009. (Abstract...) [Paper] (Copyright IEEE)

  Zernike moments constitute a powerful shape descriptor in terms of robustness and description capability. However the classical way of comparing two Zernike descriptors only takes into account the magnitude of the moments and loses the phase information. The novelty of our approach is to take advantage of the phase information in the comparison process while still preserving the invariance to rotation. This new Zernike comparator provides a more accurate similarity measure together with the optimal rotation angle between the patterns, while keeping the same complexity as the classical approach. This angle information is particularly of interest for many applications, including 3D scene understanding through images. Experiments demonstrate that our comparator outperforms the classical one in terms of similarity measure. In particular the robustness of the retrieval against noise and geometric deformation is greatly improved. Moreover, the rotation angle estimation is also more accurate than state of the art algorithms. (Hide...)

• [8] Hierarchical Watermarking of Semi-regular Meshes Based on Wavelet Transform
  Kai Wang, Guillaume Lavoué, Florence Denis and Atilla Baskurt, IEEE Transactions on Information Forensics and Security, Vol. 3, No. 4, pp. 620-634, 2008. (Abstract...) [Paper] (Copyright IEEE)

  This paper presents a hierarchical watermarking framework for semi-regular meshes. Three blind watermarks are inserted in a semi-regular mesh with different purposes: a geometrically robust watermark for copyright protection, a highcapacity watermark for carrying a large amount of auxiliary information, and a fragile watermark for content authentication. The proposed framework is based on wavelet transform of the semi-regular mesh. More precisely, the three watermarks are inserted in different appropriate resolution levels obtained by wavelet decomposition of the mesh: the robust watermark is inserted by modifying the norms of the wavelet coefficient vectors associated with the lowest resolution level; the fragile watermark is embedded in the high resolution level obtained just after one wavelet decomposition by modifying the orientations and norms of the wavelet coefficient vectors; the high-capacity watermark is inserted in one or several intermediate levels by considering groups of wavelet coefficient vector norms as watermarking primitives. Experimental results demonstrate the effectiveness of the proposed framework: the robust watermark is able to resist all the common geometric attacks even with relatively strong amplitude; the fragile watermark is robust to contentpreserving operations, while sensitive to other attacks of which it can also provide the precise location; the payload of the highcapacity watermark increases rapidly along with the number of watermarking primitives. (Hide...)

• [7] A Comprehensive Survey on Three-Dimensional Mesh Watermarking
  Kai Wang, Guillaume Lavoué, Florence Denis and Atilla Baskurt, IEEE Transactions on Multimedia, Vol. 10, No.8, pp. 1513-1527, 2008. (Abstract...) [Paper] (Copyright IEEE)

  Three-dimensional meshes have been used more and more in industrial, medical and entertainment applications during the last decade. Many researchers, from both the academic and the industrial sectors, have become aware of their intellectual property protection and authentication problems arising with their increasing use. This paper gives a comprehensive survey on 3D mesh watermarking, which is considered an effective solution to the above two emerging problems. Our survey covers an introduction to the relevant state of the art, an attack-centric investigation, and a list of existing problems and potential solutions. First, the particular difficulties encountered while applying watermarking on 3D meshes are discussed, followed by a presentation and an analysis of the existing algorithms, distinguishing them between fragile techniques and robust techniques. Since the attacks play an important role in the design of 3D mesh watermarking algorithms, we also provide an attack-centric viewpoint of this state of the art. Finally, some future working directions are pointed out especially on the ways of devising robust and blind algorithms and on some new probably promising watermarking feature spaces. . (Hide...)

• [6] Subdivision surface watermarking
  Guillaume Lavoué, Florence Denis and Florent Dupont, Computers & Graphics, Vol. 31, No.3, pp. 480-492, 2007. (Abstract...) [Paper] (Copyright Elsevier)

  This paper presents a robust non-blind watermarking scheme for subdivision surfaces. The algorithm works in the frequency domain, by modulating spectral coefficients of the subdivision control mesh. The compactness of the watermarking support (a coarse control mesh) has led us to optimize the trade-off between watermarking redundancy (which insures robustness) and imperceptibility by introducing two contributions: (1) Spectral coefficients are perturbed according to a new modulation scheme analysing the spectrum shape and (2) the redundancy is optimized by using error correcting codes coming from telecommunication theory. Since the watermarked surface can be attacked in a subdivided version, we have introduced an algorithm to retrieve the control polyhedron, starting from a subdivided, attacked version. Experiments have shown the high robustness of our scheme against geometry attacks such as noise addition, quantization or non-uniform scaling and also connectivity alterations such as remeshing or simplification. (Hide...)

• [5] A framework for quad/triangle subdivision surface fitting:  Application to mechanical objects 
  Guillaume Lavoué, Florent Dupont and Atilla Baskurt, Computer Graphics Forum, Vol. 26, No.1, pp. 1-14, 2007. (Abstract...) [Paper] (Copyright Blackwell)

  In this paper we present a new framework for subdivision surface approximation of 3D models represented by polygonal meshes. Our approach, particularly suited for mechanical or CAD parts, produces a mixed quadrangle-triangle control mesh, optimized in terms of face and vertex numbers while remaining independent of the connectivity of the input mesh. Our algorithm begins with a decomposition of the object into surface patches. Then the main idea is to approximate first the region boundaries and then the interior data. Thus, for each patch, a first step approximates the boundaries with subdivision curves (associated with control polygons) and creates an initial subdivision surface by linking the boundary control points with respect to the lines of curvature of the target surface. Then, a second step optimizes the initial subdivision surface by iteratively moving control points and enriching regions according to the error distribution. The final control mesh defining the whole model is then created assembling every local subdivision control meshes. This control polyhedron is much more compact than the original mesh and visually represents the same shape after several subdivision steps, hence it is particularly suitable for compression and visualization tasks. Experiments conducted on several mechanical models have proven the coherency and the efficiency of our algorithm, compared with existing methods. (Hide...)

• [4] High rate compression of CAD meshes based on subdivision inversion
  Guillaume Lavoué, Florent Dupont and Atilla Baskurt, Annals of Telecommunications, Vol. 60, No.11-12, pp. 1286-1310, 2005. (Abstract...) [Paper]

  In this paper we present a new framework, based on subdivision surface approximation, for efficient compression and coding of 3D models represented by polygonal meshes. Our algorithm fits the input 3D model with a piecewise smooth subdivision surface represented by a coarse control polyhedron, near optimal in terms of control points number and connectivity. Our algorithm, which remains independent of the connectivity of the input mesh, is particularly suited for meshes issued from mechanical or CAD parts. The found subdivision control polyhedron is much more compact than the original mesh and visually represents the same shape after several subdivision steps, without artifacts or cracks, like traditional lossy compression schemes. This control polyhedron is then encoded specifically to give the final compressed stream. Experiments conducted on several CAD models have proven the coherency and the efficiency of our algorithm, compared with existing methods. (Hide...)

• [3] A new subdivision based approach for piecewise smooth approximation of 3D polygonal curves
  Guillaume Lavoué, Florent Dupont and Atilla Baskurt, Pattern Recognition, Vol. 38, No.8, pp. 1139-1151, 2005. (Abstract...) [Paper] (Copyright Elsevier)

  This paper presents an algorithm dealing with the data reduction and the approximation of 3D polygonal curves. Our method is able to approximate efficiently a set of straight 3D segments or points with a piecewise smooth subdivision curve, in a near optimal way in terms of control point number. Our algorithm is a generalization for subdivision rules, including sharp vertex processing, of the Active B-Spline Curve developed by Pottmann et al. We have also developed a theoretically demonstrated approach, analysing curvature properties of B-Splines, which computes a near optimal evaluation of the initial number and positions of control points. Moreover, our original Active Footpoint Parameterization method prevents wrong matching problems occurring particularly for self-intersecting curves. Thus, the stability of the algorithm is highly increased. Our method was tested on different sets of curves and gives satisfying results regarding to approximation error, convergence speed and compression rate. This method is in line with a larger 3D CAD object compression scheme by piecewise subdivision surface approximation. The objective is to fit a subdivision surface on a target patch by first fitting its boundary with a subdivision curve whose control polygon will represent the boundary of the surface control polyhedron.(Hide...)

• [2] A new CAD mesh segmentation method, based on curvature tensor analysis
  Guillaume Lavoué, Florent Dupont and Atilla Baskurt, Computer Aided Design, Vol. 37, No. 10, pp. 975-987, 2005. (Abstract...) [Paper] (Copyright Elsevier)

  This paper presents a new and efficient algorithm for the decomposition of 3D arbitrary triangle meshes and particularly optimized triangulated CAD meshes. The algorithm is based on the curvature tensor field analysis and presents two distinct complementary steps: a region based segmentation, which is an improvement of that presented by Lavoue et al. [Lavoue G, Dupont F, Baskurt A. Constant curvature region decomposition of 3D-meshes by a mixed approach vertex-triangle, J WSCG 2004;12(2):245–52] and which decomposes the object into near constant curvature patches, and a boundary rectification based on curvature tensor directions, which corrects boundaries by suppressing their artefacts or discontinuities. Experiments conducted on various models including both CAD and natural objects, show satisfactory results. Resulting segmented patches, by virtue of their properties (homogeneous curvature, clean boundaries) are particularly adapted to computer graphics tasks like parametric or subdivision surface fitting in an adaptive compression objective. (Hide...)

• [1] Object of Interest based visual navigation, retrieval and semantic content identification system 
  Khalid Idrissi, Guillaume Lavoué, Julien Ricard and Atilla Baskurt, Computer Vision and Image Understanding, Vol. 94, No. 1-3, pp. 271-294, 2004. (Abstract...) [Paper] (Copyright Elsevier)
  

  This study presents a content-based image retrieval system IMALBUM based on local region of interest called object of interest (OOI). Each segmented or user-selected OOI is indexed with new local adapted descriptors associated to color, texture, and shape features. This local approach is an efficient way to associate the local semantic content with low-level descriptors (color, texture, shape, etc.) computed on regions selected by the user. So the user actively takes part in the indexing process (offline) and can use a selected OOI as a query for the retrieval system (online). The IMALBUM system proposes original functionalities. A visual navigation tool allows to surf in the image database when the user has no precise idea of what he is really searching for in the database. Furthermore, when an OOI is selected as a query for retrieval, a semantic content identification tool indicates to the user the probable class of this unknown object. The performance of these different tools are evaluated on different databases.(Hide...)

International conferences

• [32] Bag of Words and Local Spectral Descriptor for 3D Partial Shape Retrieval  
  Guillaume Lavoué, Eurographics Workshop on 3D Object Retrieval (3DOR), Llandudno, UK, April 2011. (Abstract...) [Paper] (Copyright Eurographics)

  This paper presents a 3D shape retrieval algorithm based on the Bag of Words (BoW) paradigm. For a given 3D shape, the proposed approach considers a set of feature points uniformly sampled on the surface and associated with local Fourier descriptors; this descriptor is computed in the neighborhood of each feature point by projecting the geometry onto the eigenvectors of the Laplace-Beltrami operator, it is highly discriminative, robust to connectivity and geometry changes and also fast to compute. In a preliminary step, a visual dictionary is built by clustering a large set of feature descriptors, then each 3D shape is described by an histogram of occurrences of these visual words. The performances of our approach have been compared against very recent state-of-theart methods on several different datasets. For global shape retrieval our approach is comparable to these recent works, however it clearly outperforms them in the case of partial shape retrieval. (Hide...)

• [31] SHREC'11 Track: Shape retrieval on Non-Rigid 3D Watertight Meshes  
  Z. Lian, A. Godil, B. Bustos, M. Daoudi, J. Hermans, S. Kawamura, Y. Kurita, G. Lavoué, H.V. Nguyen, R. Ohbuchi, Y. Ohkita, Y. Ohishi, F. Porikli, M. Reuter, I. Sipiran, D. Smeets, P. Suetens, H. Tabia, and D. Vandermeulen, Eurographics Workshop on 3D Object Retrieval (3DOR), Llandudno, UK, April 2011.  [Paper] (Copyright Eurographics)
• [30] A subjective experiment for 3D-mesh segmentation evaluation
  Halim Benhabiles, Guillaume Lavoué, Jean-Philippe Vandeborre and Mohamed Daoudi,  IEEE International Workshop on Multimedia Signal Processing (MMSP) , St-Malo, France, October 2010. (Abstract...) [Paper] (Copyright IEEE)

  In this paper we present a subjective quality assessmentexperiment for 3D-mesh segmentation. For this end, we carefully designed a protocol with respect to several factors namely the rendering conditions, the possible interactions, the rating range, and the number of human subjects. To carry out the subjective experiment, more than 40 human observers have rated a set of 250 segmentation results issued from various algorithms. The obtained Mean Opinion Scores, which represent the human subjects’ point of view toward the quality of each segmentation, have then been used to evaluate both the quality of automatic segmentation algorithms and the quality of similarity metrics used in recent mesh segmentation benchmarking systems. (Hide...)

• [29] Recognizing and localizing individual activities through graph matching  
  Anh-Phuong Ta, Christian Wolf, Guillaume Lavoué and  Atilla Baskurt, International Conference on Advanced Video and Signal-Based Surveillance (AVSS), Boston, USA, August 2010. (Abstract...) [Paper] (Copyright IEEE) Best Paper Award for the Recognition Track

  In this paper we tackle the problem of detecting individual human actions in video sequences. While the most successful methods are based on local features, which proved that they can deal with changes in background, scale and illumination, most existing methods have two main shortcomings: first, they are mainly based on the individual power of spatio-temporal interest points (STIP), and therefore ignore the spatio-temporal relationships between them. Second, these methods mainly focus on direct classification techniques to classify the human activities, as opposed to detection and localization. In order to overcome these limitations, we propose a new approach, which is based on a graph matching algorithm for activity recognition. In contrast to most previous methods which classify entire video sequences, we design a video matching method from two sets of ST-points for human activity recognition. First, points are extracted, and a hyper graphs are constructed from them, i.e. graphs with edges involving more than 2 nodes (3 in our case). The activity recognition problem is then transformed into a problem of finding instances of model graphs in the scene graph. By matching local features instead of classifying entire sequences, our method is able to detect multiple different activities which occur simultaneously in a video sequence. Experiments on two standard datasets demonstrate that our method is comparable to the existing techniques on classification, and that it can, additionally, detect and localize activities. (Hide...)

• [28] Remote scientific visualization of progressive 3D meshes with X3D  
  Adrien Maglo, Guillaume Lavoué, Céline Hudelot, Ho Lee, Christophe Mouton, Florent Dupont, International Conference on 3D Web Technology (Web3D), Los Angeles, USA, July 2010. (Abstract...) [Paper]  (Copyright ACM)

  This paper presents a framework, integrated into the X3D format, for the streaming of 3D content in the context of remote scientic visualization; a progressive mesh com- pression method is proposed that can handle 3D objects associated with attributes like colors, while producing high quality intermediate Levels Of Detail (LOD). Efficient adap- tation mechanisms are also proposed so as to optimize the LOD management of the 3D scene according to dierent constraints like the network bandwidth, the device graphic capability, the display resolution and the user preferences. Experiments demonstrate the eciency of our approach in scientic visualization scenarii. . (Hide...)

• [27] Learning an efficient and robust graph matching procedure for specific object recognition  
  Jérôme Revaud, Guillaume Lavoué, Yasuo Ariki and Atilla Baskurt, Jean-Michel Jolion, International Conference on Pattern Recognition (ICPR), Istanbul, Turkey, August 2010. (Abstract...) [Paper]  (Copyright IEEE)

  We present a fast and robust graph matching approachfor 2D specific object recognition in images. From a small number of training images, a model graph of the object to learn is automatically built. It contains its local keypoints as well as their spatial proximity relationships. Training is based on a selection of the most efficient subgraphs using the mutual information. The detection uses dynamic programming with a lattice and thus is very fast. Experiments demonstrate that the proposed method outperforms the specific object detectors of the state-of-the-art in realistic noise conditions. . (Hide...)

• [26] 3D Pairwise features for human action recognition  
  Anh-Phuong Ta, Christian Wolf, Guillaume Lavoué and  Atilla Baskurt,  International Conference on Pattern Recognition (ICPR), Istanbul, Turkey, August 2010. (Abstract...) [Paper] (Copyright IEEE)

  Existing action recognition approaches mainly relyon the discriminative power of individual local descriptors extracted from spatio-temporal interest points (STIP), while the geometric relationships among the local features1 are ignored. This paper presents new features, called pairwise features (PWF), which encode both the appearance and the spatio-temporal relations of the local features for action recognition. First STIPs are extracted, then PWFs are constructed by grouping pairs of STIPs which are both close in space and close in time. We propose a combination of two codebooks for video representation. Experiments on two standard human action datasets: the KTH dataset and the Weizmann dataset show that the proposed approach outperforms most existing methods (Hide...)

• [25] Scale-Invariant Proximity Graph for Fast Probabilistic Object Recognition  
  Jérôme Revaud, Guillaume Lavoué, Yasuo Ariki and Atilla Baskurt, ACM International Conference on Image and Video Retrieval (CIVR), Xi'an, China, July 2010.(Abstract...) [Paper] (Copyright ACM)

  A pseudo-hierarchical graph matching procedure dedicatedto object recognition is presented in this paper. From a single model image, a graph is built by extracting invariant local features and linking them according to a so-called proximity rule. The resulting graph presents several interesting properties including invariance to scale, robustness to various distortions and empirical linearity of the number of edges with respect to the number of nodes. The matching process is made hierarchical in order to increase both speed and detection performances. It relies on progressively incorporating the smaller model features as the hierarchy level increases. As a result, even a matching between graphs containing thousands of nodes is very fast (a few milliseconds). Experiments demonstrates that the method outperforms state-of-the-art specic object detectors in terms of precision-recall measures and detection time. . (Hide...)

• [24] A Benchmark for 3D Mesh Watermarking  
  Kai Wang, Guillaume Lavoué, Florence Denis,  Atilla Baskurt and Xiyan He, IEEE Shape Modeling International (SMI) , Avignon, France, June 2010. (Abstract...) [Paper] (Copyright IEEE)

  Abstract—This paper presents a benchmarkingsystem for the evaluation of robust mesh watermarking methods. The proposed benchmark has three different components: a “standard” mesh model collection, a software tool and two application-oriented evaluation protocols. The software tool integrates both geometric and perceptual measurements of the distortion induced by watermark embedding, and also the implementation of a variety of attacks on watermarked meshes. The two evaluation protocols define the main steps to follow when conducting the evaluation experiments. The efficiency of the benchmark is demonstrated through the evaluation and comparison of two recent robust algorithms. (Hide...)

• [23] A Framework for Data-Driven Progressive Mesh Compression  
  Gabriel Cirio, Guillaume Lavoué and Florent Dupont,  International Conference on Computer Graphics Theory and Applications (GRAPP),  Anger, France, May 2010. (Abstract...) [Paper]

  Progressive mesh compression techniques have reached very high compression ratios. However, these techniquesusually do not take into account associated properties of meshes such as colors or normals, no matter their size, nor do they try to improve the quality of the intermediate decompression meshes. In this work, we propose a framework that uses the associated properties of the mesh to drive the compression process, resulting in an improved quality of the intermediate decompression meshes. Based on a kd-tree geometry compression algorithm, the framework is generic enough to allow any property or set of properties to drive the compression process provided the user defines a distance function for each property. The algorithm builds the kd-tree structure using a voxelisation process, which recursively separates the set of vertices according to the associated properties distances. We evaluate our method by comparing its compression ratios to recent algorithms. In order to evaluate the visual quality of the intermediate meshes, we carried a perceptive evaluation with human subjects. Results show that at equal rates, our method delivers an overall better visual quality. The algorithm is particularly well suited for the compression of meshes where geometry and topology play a secondary role compared to associated properties, such as with many scientific visualization models. (Hide...)

• [22] New methods for progressive compression of colored 3D Mesh  
  Ho Lee, Guillaume Lavoué and Florent Dupont,  International Conference on Computer Graphics, Visualization and Computer Vision (WSCG), Plzen, Czech Republic, February 2010. (Abstract...) [Paper]

  In this paper, we present two methods to compress colored 3D triangular meshes in a progressive way. Althoughmany progressive algorithms exist for efficient encoding of connectivity and geometry, none of these techniques consider the color data in spite of its considerable size. Based on the powerful progressive algorithm from Alliez and Desbrun [All01a], we propose two extensions for progressive encoding and reconstruction of vertex colors: a prediction-based method and a mapping table method. In the first one, after transforming the initial RGB space into the Lab space, each vertex color is predicted by a specific scheme using information of its neighboring vertices. The second method considers a mapping table with reduced number of possible colors in order to improve the rate-distortion tradeoff. Results show that the prediction method produces quite good results even in low resolutions, while the mapping table method delivers similar visual results but with a fewer amount of bits transmitted depending on the color complexity of the model. (Hide...)

• [21] Adaptive coarse-to-fine quantization for optimizing rate-distortion of progressive mesh compression  
  Ho Lee, Guillaume Lavoué and Florent Dupont,  Vision, Modeling, and Visualization Workshop (VMV), Braunschweig, Germany, November 2009. (Abstract...) [Paper]

  We propose a new connectivity-based progressive compression approach for triangle meshes. The key idea is to adapt the quantization precision to the resolution of each intermediate mesh so as to optimize the rate-distortion trade-off. This adaptation is automatically determined during the encoding process and the overhead is efficiently encoded using geometrical prediction techniques. We also introduce an optimization of the geometry coding by using a bijective discrete rotation. Results show that our approach delivers a better rate-distortion behavior than both connectivity-based and geometry-based compression state of the art methods. (Hide...)

• [20]  Global triangular mesh regularization using conditional Markov random fields
  Vincent Vidal, Christian Wolf, Florent Dupont, Guillaume Lavoué, Symposium on Geometry Processing (SGP), Poster, Berlin, Germany, July 2009. [Poster] 
• [19] Local patch blind spectral watermarking method for 3D graphics  
  Ming Luo, Kai Wang, Adrian G. Bors and  Guillaume Lavoué,  International Workshop on Digital Watermarking (IWDW), Lecture Notes on Computer Science, Guildford, UK, August 2009. (Abstract...) [Paper] (Copyright Springer)

  In this paper, we propose a blind watermarking algorithm for 3D meshes. The proposed algorithm embeds spectral domain constraints in segmented patches. After aligning the 3D object using volumetric moments , the patches are extracted using a robust segmentation method which ensures that they have equal areas. One bit is then embedded in each patch by enforcing specific constraints in the distribution of its spectral coefficients by using Principal Component Analysis (PCA). A series of experiments and comparisons with state-of-the-art in 3D graphics wa- termarking have been performed; they show that the proposed scheme provides a very good robustness against both geometry and connectivity attacks, while introducing a low level of dis torsion. (Hide...)

• [18] 3D Object detection and viewpoint selection in sketch images using local patch-based Zernike moments  
  Anh-Phuong Ta, Christian Wolf, Guillaume Lavoué and  Atilla Baskurt,  IEEE Workshop on Content Based Multimedia Indexing (CBMI), Crete, Greece, June 2009. (Abstract...) [Paper] (Copyright IEEE)

  In this paper we present a new approach to detect and recognize 3D models in 2D storyboards which have been drawn during the production process of animated cartoons. Our method is robust to occlusion, scale and rotation. The lack of texture and color makes it difficult to extract local features of the target object from the sketched storyboard. Therefore the existing approaches using local descriptors like interest points can fail in such images. We propose a new framework which combines patch-based Zernike descriptors with a method enforcing spatial constraints for exactly detecting 3D models represented as a set of 2D views in the storyboards. Experimental results show that the proposed method can deal with partial object occlusion and is suitable for poorly textured objects. (Hide...)

• [17] A framework for the objective evaluation of segmentation algorithms using a ground-truth of human segmented 3D-models
  Halim Benhabiles, Jean-Philippe Vandeborre, Guillaume Lavoué and Mohamed Daoudi,  IEEE Shape Modeling International (SMI), Beijing, China, June  2009. (Abstract...) [Paper] (Copyright IEEE)

  In this paper, we present an evaluation method of 3D-mesh segmentation algorithms based on a ground-truth corpus. This corpus is composed of a set of 3D-models grouped in different classes (animals, furnitures, etc.) associated with several manual segmentations produced by human observers. We define a measure that quantifies the consistency between two segmentations of a 3D-model, whatever their granularity. Finally, we propose an objective quality score for the automatic evaluation of 3D-mesh segmentation algorithms based on these measures and on the ground-truth corpus. Thus the quality of segmentations obtained by automatic algorithms is evaluated in a quantitative way thanks to the quality score, and on an objective basis thanks to the ground-truth corpus. Our approach is illustrated through the evaluation of two recent 3D-mesh segmentation methods. (Hide...)

• [16] Markov Random Fields for Improving 3D Mesh Analysis and Segmentation  
  Guillaume Lavoué and Christian Wolf, Eurographics 2008 Workshop on 3D Object Retrieval, pp. 25-32, Crete, Greece, April 2008. (Abstract...) [exe] [Paper] (Copyright Eurographics)

  Mesh analysis and clustering have became important issues in order to improve the efficiency of common processingoperations like compression, watermarking or simplification. In this context we present a new method for clustering / labeling a 3D mesh given any field of scalar values associated with its vertices (curvature, density, roughness etc.). Our algorithm is based on Markov Random Fields, graphical probabilistic models. This Bayesian framework allows (1) to integrate both the attributes and the geometry in the clustering, and (2) to obtain an optimal global solution using only local interactions, due to the Markov property of the random field. We have defined new observation and prior models for 3D meshes, adapted from image processing which achieve very good results in terms of spatial coherency of the labeling. All model parameters are estimated, resulting in a fully automatic process (the only required parameter is the number of clusters) which works in reasonable time (several seconds). (Hide...)

• [15] A Fragile Watermarking Scheme for Authentication of Semi-regular Meshes  
  Kai Wang, Guillaume Lavoué, Florence Denis and Atilla Baskurt, Eurographics 2008, Short Paper, pp. 5-8, Crete, Greece, April 2008. (Abstract...) [Paper] (Copyright Eurographics)

  This paper presents a fragile watermarking scheme for authentication of 3D semi-regular meshes. After onewavelet decomposition, the watermark is inserted by slightly modifying the norms and orientations of the obtained wavelet coefficient vectors. The inserted watermark is robust to the so-called content-preserving attacks including vertex reordering and similarity transformations. However, it is vulnerable to others attacks such as local and global geometric modifications and remeshing since the objective is to check the integrity of the mesh. Additionally, according to the watermark extraction result, these attacks can be precisely located on the surface of the attacked mesh in a blind way. Sufficient security level is also achieved by introducing secret keys and by using scalar Costa quantization scheme with appropriate parameter values. Experimental results demonstrate the efficacy of the proposed watermarking scheme. (Hide...)

• [14] Feature-Based Subdivision Surface Fitting  
  Guillaume Lavoué and Florent Dupont, ACM SIGGRAPH 2007, Poster, San Diego, California, USA, August 2007. [Abstract paper] (Copyright ACM)
• [13] A Roughness Measure for 3D Mesh Visual Masking  
  Guillaume Lavoué, ACM SIGGRAPH Symposium on Applied Perception in Graphics and Visualization (APVG), pp. 57-60, Tübingen, Germany, July 2007.(Abstract...) [exe] [Paper] (Copyright ACM)

  3D models are subject to a wide variety of processing operations such as compression, simplification or watermarking, which introduce slight geometric modifications on the shape. The main issue is to maximize the compression/simplification ratio or the watermark strength while minimizing these visual degradations. However few algorithms exploit the human visual system to \textit{hide} these degradations, while perceptual attributes could be quite relevant for this task. Particularly, the Masking Effect defines the fact that a signal can be masked by the presence of another signal with similar frequency or orientation. In this context we introduce the notion of \textit{roughness} for a 3D mesh, as a local measure of geometric noise on the surface. Indeed, a textured (or \textit{rough}) region is able to hide geometric distortions much better than a smooth one. Our measure is based on curvature analysis on local windows of the mesh and is independent of the resolution/connectivity of the object. An application to Visual Masking is presented and discussed. (Hide...)

• [12] Fast and cheap object recognition by linear combination of views  
  Jérôme Revaud, Guillaume Lavoué, Yasuo Ariki and Atilla Baskurt, ACM International Conference on Image and Video Retrieval (CIVR), pp. 194-201, Amsterdam, The Netherlands, July 2007.(Abstract...) [Paper] (Copyright ACM)

  In this paper, we present a real-time algorithm for 3D object detection in images. Our method relies on the Ullman and Basri theory which claims that the same object under different transformations can often be expressed as the linear combinations of a small number of its views. Thus, in our framework the 3D object is modelized by two 2D images associated with spatial relationships described by localinvariant feature points. The recognition is based on feature points detection and alignment with the model. Important theoretical optimizations have been introduced in order to speed up the original full alignment scheme and to reduce the model size in memory. The recognition process is based on a very fast recognition loop which quickly eliminates outliers. The proposed approach does not require a segmentation stage, and it is applicable to cluttered scenes. The small size of the model and the rapidity of the detection make this algorithm articularly suitable for real-time applications on mobile devices. (Hide...)

• [11] Hierarchical Blind Watermarking of 3D Triangular Meshes  
  Kai Wang, Guillaume Lavoué, Florence Denis and Atilla Baskurt, IEEE International Conference on Multimedia & Expo (ICME), pp. 1235-1238, Beijing, China, July 2007. (Abstract...) [Preprint] (Copyright IEEE)

  An original hierarchical watermarking scheme is proposed in this paper. A geometrically robust watermark and a high-capacity watermark are inserted in different resolution levels of the wavelet decomposition of a semi-regular mesh by modifying the norms of wavelet coefficients. Both watermarks are blind and invariant to similarity transformations. The robustness of the first watermark is achieved by synchronizing and quantizing watermark primitives according to edges lengths of the coarsest level, which are quite insensible to geometrical attacks. The high capacity of the second watermark is obtained by considering the permutation of the norms of a group of wavelet coefficients. Experiments have proven the high robustness of the first watermark under common geometrical attacks. To our knowledge, the capacity of the second method, which can attain the factorial of the candidate coefficients number, is the highest for 3D meshes in the literature. (Hide...)

• [10] Three-Dimensional Meshes Watermarking: Review and Attack-Centric Investigation  
  Kai Wang, Guillaume Lavoué, Florence Denis and Atilla Baskurt, Information Hiding (IH), Lecture Notes on Computer Science, Vol. 4567, pp. 50-64, Saint Malo, France, June 2007. (Abstract...) [Preprint] (Copyright Springer)

  The recent decade has seen the emergence of 3D meshes in industrial, medical and entertainment applications. Therefore, their intellectual property protection problem has attracted more and more attention in both the research and industrial realms. This paper gives a synthetic review of 3D mesh watermarking techniques, which are deemed to be a potential effective solution to the above problem. We begin with a discussion on the particular difficulties encountered in applying watermarking on 3D meshes. Then some typical algorithms are presented and analyzed, classifying them in two categories: spatial and spectral. Considering the important impact of the different attacks on the design of 3D mesh watermarking algorithms, we provide an attack-centric viewpoint of this state of the art. Finally, some special issues and possible future working directions are discussed. (Hide...)

• [9] A Watermarking Framework for Subdivision Surfaces 
  Guillaume Lavoué, Florence Denis, Florent Dupont and Atilla Baskurt, International Workshop on Multimedia Content Representation, Classification and Security (IWMRCS), Lecture Notes on Computer Science, Vol. 4105, pp. 223-231, Istanbul, Turkey, September 2006. (Abstract...) [Paper] (Copyright Springer)

  This paper presents a robust watermarking scheme for 3D subdivision surfaces. Our proposal is based on a frequency domain decomposition of the subdivision control mesh and on spectral coefficients modulation. The compactness of the cover object (the coarse control mesh) has led us to optimize the trade-off between watermarking redundancy (which insures robustness) and imperceptibility by introducing two contributions: (1) Spectral coefficients are perturbed according to a new modulation scheme analyzing the spectrum shape and (2) the redundancy is optimized by using error correcting codes. Since the watermarked surface can be attacked in a subdivided version, we have introduced a so-called synchronization algorithm to retrieve the control polyhedron, starting from a subdivided, attacked version. Through the experiments, we have demonstrated the high robustness of our scheme against both geometry and connectivity alterations. (Hide...)

• [8] Perceptually driven 3D distance metrics with application to watermarking 
  Guillaume Lavoué, Elisa Drelie Gelasca, Florent Dupont, Atilla Baskurt and Touradj Ebrahimi, SPIE Applications of Digital Image Processing XXIX, San Diego, August 2006. (Abstract...)  [exe] [Paper] (Copyright SPIE)

  This paper presents an objective structural distorsion measure which reflects the visual similarity between 3D meshes and thus can be used for quality assessment. The proposed tool is not linked to any specific application and thus can be used to evaluate any kinds of 3D mesh processing algorithms (simplification, compression, watermarking etc.). This measure follows the concept of structural similarity recently introduced for 2D image quality assessment by Wang \emph{et al.} \cite{WANG:2004} and is based on curvature analysis (mean, standard deviation, covariance) on local windows of the meshes. Evaluation and comparison with geometric metrics are done through a subjective experiment based on human evaluation of a set of distorted objects. A quantitative perceptual metric is also derived from the proposed structural distorsion measure, for the specific case of watermarking quality assessment, and is compared with recent state of the art algorithms. Both visual and quantitative results demonstrate the robustness of our approach and its strong correlation with subjective ratings. (Hide...)

• [7] Digital Watermarking of Compressed 3D Meshes
  Florence Denis, Guillaume Lavoué, Florent Dupont and Atilla Baskurt, International Conference on Machine Intelligence (ACIDCA-ICMI’2005), Tozeur, Tunisia, November, 2005. (Abstract...) [Paper]

  This paper presents a robust watermarking algorithm applied to 3D compressed polygonal meshes. Copyright protection of 3D models becomes very important for many applications using public networks. As some recent compression techniques allow very high compression rates, it is of interest to verify that watermarking techniques support this kind of attack. In this paper we present the complete scheme developed: the compression algorithm, the watermarking algorithm and the mark extraction process. (Hide...)

• [6] High rate compression of 3D meshes using a subdivision scheme
  Guillaume Lavoué, Florent Dupont and Atilla Baskurt, European Signal Processing Conference (EUSIPCO’2005), Antalya, Turkey, September, 2005. [Paper]
• [5] Subdivision surface fitting for efficient compression and coding of 3D models
  Guillaume Lavoué, Florent Dupont and Atilla Baskurt, SPIE Visual Communications and Image Processing (VCIP'2005), Vol. 5960, pp. 1159-1170, Beijing, China, July, 2005. (Abstract...) [Paper] (Copyright SPIE)

  In this paper we present a new framework, based on subdivision surface fitting, for high rate compression and coding of 3D models. Our algorithm fits the input 3D model, represented by a polygonal mesh, with a piecewise smooth subdivision surface represented by a coarse control polyhedron. Our fitting scheme, particularly suited for meshes issued from mechanical or CAD parts, aims at getting close to the optimality in terms of control points number, while remaining independent of the connectivity of the input mesh. The found subdivision control polyhedron is much more compact than the original mesh and visually represents the same shape after several subdivision steps, without artifacts or cracks, like traditional lossy compression schemes. This control polyhedron is then encoded specifically to give the final compressed stream. Experiments conducted on several 3D models have proven the coherency and the efficiency of our framework, compared with existing compression methods.(Hide...)

• [4] Toward a near optimal quad/triangle subdivision surface fitting
  Guillaume Lavoué, Florent Dupont andAtilla Baskurt, IEEE Int. Conf. on 3-D Digital Imaging and Modeling (3DIM'2005), pp. 402-409, Ottawa, Canada, June 2005. (Abstract...) [Paper] (Copyright IEEE Computer Society)

  In this paper we present a new framework for subdivision surface fitting of arbitrary surfaces (not closed objects) represented by polygonal meshes. Our approach is particularly suited for output surfaces from a mechanical or CAD object segmentation for a piecewise subdivision surface approximation. Our algorithm produces a mixed quadrangle-triangle control mesh, near optimal in terms of face and vertex numbers while remaining independent of the connectivity of the input mesh. The first step approximates the boundaries with subdivision curves and creates an initial subdivision surface by optimally linking the boundary control points with respect to the lines of curvature of the target surface. Then, a second step optimizes the initial control polyhedron by iteratively moving control points and enriching regions according to the error distribution. Experiments conducted on several surfaces and on a whole segmented mechanical object, have proven the coherency and the efficiency of our algorithm, compared with existing methods.(Hide...)

• [3] Curvature tensor Based Triangle Mesh Segmentation with Boundary Rectification
  Guillaume Lavoué, Florent Dupont and Atilla Baskurt, IEEE Computer Graphics International (CGI'2004), pp. 10-17, Crete, Greece, June 2004. (Abstract...) [Paper] (Copyright IEEE Computer Society)

  This paper presents a new and efficient algorithm for the decomposition of 3D arbitrary triangle meshes into surface patches. The algorithm is based on the curvature tensor field analysis and presents two distinct complementary steps: a region based segmentation, which is an improvement of that presented in Lavoue et al. [Lavoue G, Dupont F, Baskurt A. Constant curvature region decomposition of 3D-meshes by a mixed approach vertex-triangle, J WSCG 2004;12(2):245–52] and which decomposes the object into known and near constant curvature patches, and a boundary rectification based on curvature tensor directions, which corrects boundaries by suppressing their artifacts or discontinuities. Experiments were conducted on various models including both CAD and natural objects, results are satisfactory. Resulting segmented patches, by virtue of their properties (known curvature, clean boundaries) are particularly adapted to computer graphics tasks like parametric or subdivision surface fitting in an adaptive compression objective. (Hide...)

• [2] Constant Curvature Region Decomposition of 3D-Meshes by a Mixed Approach Vertex-Triangle
  Guillaume Lavoué, Florent Dupont and Atilla Baskurt, Journal of WSCG (WSCG’2004), Vol.12, No.2, pp. 245-252, ISSN 1213-6972, February 2-6, 2004, Plzen, Czech Republic. (Abstract...) [Paper]

  We present a new and efficient algorithm for decomposition of arbitrary triangle meshes into connected subsets of meshes called regions. Our method, based on discrete curvature analysis decomposes the object into almost constant curvature surfaces and not only “cut” the object along its hard edges like traditional methods. This algorithm is an hybrid approach vertex-triangle, it is based on three major steps: vertices are first classified using their discrete curvature values, then connected triangle regions are extracted via a region growing process and finally similar regions are merged using a region adjacency graph in order to obtain final patches. Experiments were conducted on both CAD and natural models, results are satisfactory. Segmented patches can then be used instead of the complete complex model to facilitate computer graphic tasks such as smoothing, surface fitting or compression. (Hide...)

• [1] Système de Recherche d'Images et de Navigation Visuelle Basé sur une Approche Locale
  Khalid Idrissi, Guillaume Lavoué, Julien Ricard, Int. Conf. on Image and Signal Processing (ICISP'2003), pp. 119-128, Agadir, Maroc, June 2003.

National conferences

• J. Revaud, Yasuo Ariki, G. Lavoué, A. Baskurt, Combinaison de caractéristiques pour la reconnaissance rapide, robuste etinvariante d’objets spécifiques, Proc. of Colloque RFIA, Caen, France, 2010.
• H. Benhabiles, G. Lavoué, J.P. Vandeborre, M. Daoudi, Une nouvelle métrique de similarité pour l’évaluation desegmentations de maillages 3D en utilisant un corpus de vérités-terrains , Journées de l'AFIG, Arles, France, 2009.
• J. Revaud, G. Lavoué, A. Baskurt, Une nouvelle mesure de distance entre descripteurs de moments de Zernike pour une similarité optimale et un angle de rotation entre les images , Proc. of Colloque CORESA'09, Toulouse, France, 2009.
• H. Benhabiles, J.P. Vandeborre, G. Lavoué, M. Daoudi, Une collection de modèles 3D avec vérité-terrain pour l’évaluation objective des algorithmes de segmentation , Proc. of Colloque CORESA'09, Toulouse, France, 2009.
• K. Wang, G. Lavoué, F. Denis, A. Baskurt, Tatouage robuste et aveugle de maillages 3D basé sur les moments volumiques, Proc. of Colloque CORESA'09, Toulouse, France, 2009.
• K. Wang, G. Lavoué, F. Denis, A. Baskurt, Tatouage hiérarchique et aveugle de maillages tridimensionnels, Proc. of Colloque CORESA'07, Montpellier, France, 2007.
• Lavoué, G., Dupont, F., Baskurt, A., Rectification directionnelle de frontières pour la segmentation de maillages, Proc. of Colloque CORESA'04, Lille, France, May 2004. [Paper]
• A. Baskurt, F. Dupont, G. Lavoué, J. Ricard, M. Daoudi, J.P. Vandeborre, T. Filali, J.L. Dugelay, E. Garcia, F. Preteux, T. Zaharia, F. Blum, B. Savage, D. Lichau, F. Fratani, A. Dutartre, SEMANTIC-3D : compression, indexation et tataouge de données 3D, Proc. of Colloque CORESA'04, Lille, France, May 2004. [Paper]
• Lavoué, G., Idrissi, K., Baskurt, A., Accès aux bases d’images par navigation visuelle, Proc. of Colloque CORESA'03, pp. 55-58, Lyon, France, January 2003. [Paper]

PhD Thesis

• Lavoué, G.,  Compression de surfaces, basée sur la subdivision inverse, pour la transmission bas débit et la visualisation progressive, Thèse soutenue le 1er Décembre 2005, à l'université Claude Bernard, Lyon 1. [PDF]