Simultaneous Motion Detection and Background Reconstruction with a Mixed-State Conditional Markov Random Field

Abstract

We consider the problem of motion detection by background subtraction. An accurate estimation of the background is only possible if we locate the moving objects; meanwhile, a correct motion detection is achieved if we have a good available background model. This work proposes a new direction in the way such problems are considered. The main idea is to formulate this class of problem as a joint decision-estimation unique step. The goal is to exploit the way two processes interact, even if they are of a dissimilar nature (symbolic-continuous), by means of a recently introduced framework called mixed-state Markov random fields. In this paper, we will describe the theory behind such a novel statistical framework, that subsequently will allows us to formulate the specific joint problem of motion detection and background reconstruction. Experiments on real sequences and comparisons with existing methods will give a significant support to our approach. Further implications for video sequence inpainting will be also discussed.

Multiscale NeighborhoodWise Decision Fusion for Redundancy Detection in Image Pairs

“…The potential of our method shown in Fig. 7.14 is satisfying when compared to the state-of-the-art background subtraction methods which can handle illumination changes and moving objects [100,41,93,72,109,33,46,79]. Our method did the same job using two input images and sometimes outperformed several methods [93,41,2,46,79] which require a long image sequence as input.…”

“…Unlike previous methods (kernelbased [41,72,92] or mixture of Gaussian-based [93,109,51]) which assume a time series of images as input, we consider the scenario where we have in our possession two images, without prior knowledge of the scene as considered in [65]. Background subtraction based on temporal information cannot be performed as described in [41,93,31,96,33] since we are using two images only.…”

“…We compare our approach to commonly-used and more recent background subtraction methods [93,41,109,33,79,26] applied to images taken from video-surveillance image sequences. All these change detection methods exploit several frames of the sequence to determine the foreground objects unlike our multiscale detection method.…”

Dynamic Scene Classification: Learning Motion Descriptors with Slow Features Analysis

“…The detailed results 6 are presented in tables 1 and 2. Similar conclusions can be drawn from both datasets.…”

“…In order to explicitly model texture dynamics, linear dynamical systems (LDS) have been proposed in [32]. Such stochastic models have been successfully applied in various contexts, from dynamic texture classification to motion segmentation [6] or tacking [5]. However, LDS is intrinsically limited by the first-order markov property and linearity assumption.…”