Institute for Intelligent Systems :: Video Image Segmentation, Tracking and Modeling


Video Image Segmentation, Tracking and Modeling
The objective of the project is to segment, detect, track and model an object or multiple objects in video images using a Hidden Markov Model (HMM). A spatio-temporal segmentation algorithm is developed to segment each image into a set of regions. A shot cut boundary detection algorithm is used to break the video sequence into a set of shots. An adjustable frame rate is used to track object movements over time with a pre-defined set of states based on position. The model will be trained to detect motion based on the position of the object, whether it is approaching or moving away from the camera. Various applications include advanced automatic surveillance videos, and automatic recognition systems among others.	Publications Contact Us

Last updated: 05/20/2008 16:24:38