作者: A.M. Baumberg , D.C. Hogg
DOI: 10.1109/MNRAO.1994.346236
关键词: Real image 、 Spline (mathematics) 、 Pattern recognition 、 Silhouette 、 Artificial intelligence 、 Computer vision 、 Computer science 、 Kalman filter 、 Active contour model 、 Automatic control 、 Active shape model 、 Iterative refinement
摘要: There has been considerable research interest recently, in the areas of real time contour tracking and active shape models. This paper demonstrates how dynamic filtering can be used combination with a modal-based flexible model to track an articulated non-rigid body motion. The results show method being silhouette walking pedestrian time. was generated automatically from image data incorporates variability due orientation as well object flexibility. A Kalman filter is control spatial scale for feature search over successive frames. Iterative refinement allows accurate localisation where feasible. knowledge likely speeds up by reducing number system parameters. further increase speed obtained parameters independently. >