Fuzzy dynamic model for feature tracking

Feature tracking is one of the most challenging and important tasks in Motion Analysis which plays an important role in several areas of Computer Vision. In this work, a novel approach for feature tracking based on Fuzzy concepts is introduced. Fuzzy Sets related with both cinematic (movement model) and non cinematic (image gray levels) properties are constructed in order to model the feature motion. Meanwhile cinematic related fuzzy sets model the feature movement characteristics, the non cinematic fuzzy sets model the feature visible image related properties. The final motion model is obtained through the fusion of these fuzzy models by means of a fuzzy inference engine. Experimental results are presented showing that the approach successfully copes with usual difficulties within this problem.