Spatiotemporal Autoregressive Research Articles

A new image prediction model based on spatio-temporal techniques

This paper addresses an image prediction problem focused on images with no identifiable objects. In it, we present several approaches to predict the next image of a given sequence, when the image lacks the well-defined objects, such as meteorological maps or satellite imagery. In these images no clear borders are present, and any object candidate moves, changes, appears and disappears in any image. Nevertheless, this evolution, though unrestricted, is gradual and, hence, prediction looks feasible. One of the approaches presented here, based on a spatio-temporal autoregressive (STAR) model, offers good results for these kinds of images. The main contribution of this paper is to adapt spatio-temporal models to an image prediction problem. As a byproduct of this research, we have achieved a new image compression method, suitable for images without defined shapes.

Using Temporal Texture for Content-Based Video Retrieval

Textures evolving over time are called temporal textures and are very common in everyday life. Examples are the smoke flowing or the wavy water of a river. The idea explored in this paper is that image features based on temporal texture could allow a better performance of current content-based video retrieval systems that are mainly based on static characteristics of representative frames, like color and texture. To this aim we analyze the spatio-temporal nature of texture and its application in content-based access to video databases. In particular, we represent temporal texture using the spatio-temporal autoregressive (STAR) model and a variation of self-organizing maps (SOM) where each node is an autoregressive model. These representation schemes have been implemented in a query by example framework to analyze the weaknesses and the strengths of the different approaches. Preliminary experimental results are reported.