Brahim AKSASSE

This paper deals with the three-dimensional Autoregressive (3-D AR) model parameter estimation from noisy data. We develop an algorithm to estimate the transversal AR parame- ters corresponding to the Quarter-Space (QS) region of sup- port without a priori knowledge of additive noise power. The transversal parameters and the noise variance are both ob- tained as a solution of a quadratic

Recent scanning technology and 3D modeling allowed having a large 3D meshes database. These models are widely used in several areas such as CAD, computer graphics and audiovisual production. Content based retrieval is a necessary solution to structure, to manage the multimedia data, and to navigate in these databases. In this paper, we propose a method to automatically search and retrieve 3D models visually similar to a query 3D model. This is based on the representation of a 3D model by a series of slices along a direction; the nearest models to the query are those which have cuts similar to it.