Analysis Of Complex Scenes Research Articles

Boundary-Control Vector (BCV) Motion Field Representation and Estimation by Using a Markov Random Field Model

A new motion field representation based on the boundary-control vector (BCV) scheme for video coding is examined in this work. With this scheme, the motion field is characterized by a set of control vectors and boundary functions. The control vectors are associated with the center points of blocks to control the overall motion behavior. We use the boundary functions to specify the continuity of the motion field across adjacent blocks. For BCV-based motion field estimation, an optimization framework based on the Markov random field model and maximuma posteriori(MAP) criterion is used. The new scheme effectively represents complex motions such as translation, rotation, zooming, and deformation and does not require complex scene analysis. Compared with MPEG of similar decoded SNR (signal-to-noise ratio) quality, 15–65% bit rate saving can be achieved in the proposed scheme with a more pleasant visual quality.

Detection and description of moving objects by stochastic modelling and analysis of complex scenes

Abstract This paper presents a new technique for the detection and description of moving objects in natural scenes which is based on a statistical multi-feature analysis of video sequences. In most conventional schemes for the detection of moving objects, temporal differences of subsequent images from a video sequence are evaluated by so-called change detection algorithms. These methods are based on the assumption that significant temporal changes of an image signal are caused by moving objects in the scene. However, as temporal changes of an image signal can as well be caused by many other sources (camera noise, varying illumination, small camera motion), such systems are afflicted with the dilemma of either causing many false alarms or failing to detect relevant events. To cope with this problem, the additional features of texture and motion beyond temporal signal differences are extracted and evaluated in the new algorithm. The adaptation of this method to normal fluctuations of the observed scene is performed by a time-recursive space-variant estimation of the temporal probability distributions of the different features (signal difference, texture and motion). Feature data which differ significantly from the estimated distributions are interpreted to be caused by moving objects.