Abstract. Dynamic, or temporal, texture is a spatially repetitive, timevarying visual pattern that forms an image sequence with a certain temporal stationarity. Important tasks are thus the detection, segmentation and perceptual characterization of Dynamic Texture (DT). Following recent work, color image decomposition appears as a good way to reach these different aims, however, to our best knowledge, no proposed model is currently able to deal with temporal aspect, inherent to color image sequences.The major contribution of this paper is to adapt static decomposition model to time aspect in order to deal with videos and color image sequences. In this paper we propose an extended decomposition model which splits a video into two components, a first one containing geometrical information, the structure of the sequence and a second one dynamic color texture and noise. Examples for color video decomposition and characterization of real dynamic present in texture component will be presented.

Abstract. According to recent works, introduced by Y.Meyer [1] the decomposition models based on Total Variation (TV) appear as a very good way to extract texture from image sequences. Indeed, videos show up characteristic variations along the temporal dimension which can be catched in the decomposition framework. However, there are very few works in literature which deal with spatio-temporal decompositions. Thus, we devote this paper to spatio-temporal extension of the spatial color decomposition model. We provide a relevant method to accurately catch Dynamic Textures (DT) present in videos. Moreover, we obtain the spatio-temporal regularized part (the geometrical component), and we distinctly separate the highly oscillatory variations, (the noise). Furthermore, we present some elements of comparison between several models in denoising purpose.