Mesh Processing

Abstract

It is well-known that semi-regular meshes have a relevant structure for several applications, like compression or LOD. Therefore we are developing several applications or tools for this kind of meshes.

Selected Relative Publications

Optimized Butterfly-based lifting scheme for semi-regular meshes A. Kammoun, F. PAYAN, M. Antonini IEEE International Conference in Image Processing (ICIP), September, 2011.

Adaptive Semi-Regular Remeshing: A Voronoi-Based Approach. A. Kammoun, F. PAYAN, M. Antonini In Proceedings of IEEE international workshop on MultiMedia Signal Processing (MMSP), Saint-Malo, France, September, 2010.

Measuring Errors for Massive Triangle Meshes. A. Meftah, A. Roquel, F. PAYAN, M. Antonini In Proceedings of IEEE international workshop on MultiMedia Signal Processing (MMSP), Saint-Malo, France, September, 2010.

Animated Model Compression

Abstract

During these works, we present an efficient compression scheme for animated sequences of triangular meshes of the same connectivity. The proposed algorithm exploits the temporal coherence of the geometry component by using a temporal wavelet filtering. The quantization of the resulting wavelet coefficients is then optimized by a bit allocation process. This process dispatches the bit budget across the coefficient subbands according to their influence on the quality of the reconstructed sequence for one specific user-given bitrate. The proposed scheme is simple, fast, flexible, and scalable in frame rate and bitrate. Moreover, simulation results show that our approach is competitive for any kind of animated models, whatever the characteristics (parametrically coherent or not, fine/coarse meshes...), contrary to the related works.

Selected Relative Publications

Remeshing and Spatio-Temporal Wavelet Filtering for 3D Animations Frédéric Payan, Aymen Kamoun, Marc Antonini Proceedings of IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), Las Vegas, March-April, 2008.

Temporal Wavelet-based Geometry Coder for 3D Animations [PREPRINT] F. Payan, M. Antonini Computer & Graphics, Elsevier, vol. 31(1), Jan. 2007.

Multiresolution Modeling of Dynamic Meshes

Problem

Although deforming surfaces are frequently used in numerous domains (scientific applications, games...), only few works have been proposed until now for simplifying such data. However, these time-varying surfaces are generally represented as oversampled triangular meshes with a static connectivity, involving a large number of unnecessary details for some frames. Among the related works, some methods provide good results, but fine details appearing during the animation are not always well-preserved. The objective of my current work is to propose a simplification methods preserving the fine details appearing during the animation.

Ces travaux ont été effectués durant mon post-doctorat CNRS au sein de l'équipe Modélisation Géométrique et Approximation du Laboratoire de Modélisation et de Calcul (IMAG), à Grenoble. Mon ancienne page web est donc ici !

Selected Relative Publications

Deforming surface simplification based on dynamic geometry sampling F. Payan, Stefanie Hahmann, Georges-Pierre Bonneau Proceedings of IEEE Shape Modeling International (SMI), Lyon, France, June, 2007.

Results

Les details fins sont préservés (doigts, oreilles, etc.).

Geometry Compression (PhD Degree)

Abstract

the context of this work is the rate-distortion optimization for triangular meshes quantized by a geometry coder. Some works consider that geometry is the most important component of polygonal meshes. Geometry coders include a remeshing technique and a wavelet transform providing an efficient multiresolution analysis. The goal of our works is to propose a geometry coder including a bit allocation process. This process would optimize the visual quality of reconstructed meshes for a specific bitrate. The bit allocation aims to minimize a "surface-to-surface distance" between the input mesh and the quantized one according to a target bitrate. This distance which captures the geometric differences betweeen two surfaces is a computationally intensive process. An approximation is thus a good way to allow a fast allocation. Therefore we have shown that under some assumptions, this distance can be approximated by a weighted sum of quantization errors related to the wavelet coefficients. Moreover, the statistical properties of the wavelet coefficients allow to use theoretical models for the bitrate and the distortion. Finally, we propose an efficient geometry coder including a fast bit allocation optimizing the quantization of the coefficients. This allocation improves the visual quality of the reconstructed object according to a user-given bitrate. Experimentally, the proposed coder provides better results than state-of-the-art methods.

Selected Relative Publications

Mean Square Error Approximation for Wavelet-based Semiregular Mesh Compression F. Payan, M. Antonini In IEEE Transactions on Visualization and Computer Graphics (TVCG), July/August 2006.

An Efficient Bit Allocation for Compressing Normal Meshes with an Error-driven Quantization F. Payan, M. Antonini In a Special issue on Geometric Mesh Processing of Elsevier Computer Aided Geometric Design, July 2005.

PhD Degree (french manuscript, english slides): Optimisation du compromis débit-distorsion pour la compression géométrique de maillages surfaciques triangulaires F. Payan supervisor: Marc Antonini (CReATIVe Research Group), presented in december 2004.