Dynamic Object Tracking and Occlusion Detection Based on Extended and Advanced Approach

In general Occlusion Detection is a challenging problem. Although different Detection algorithms were proposed, they all have problems in detecting occluded Objects. Some of those are Random Subspace Method (RSM), Support Vector Machine (SVM), Active Contour Model.RSM Approach is used to detect the Pedestrians. To implement these mainly three types of datasets are to be taken PobleSec, INRIA, and Daimler Multicue dataset, Aditionally used linear SVM for Classification. Active Contour Model is used for finding object outline from an image. This Approach takes the advantage of the point distribution model to limit the shape. This algorithm is highly sensitive to the initialization of tracking, making it difficult to start tracking automatically. So to overcome this problem, in this project a novel technique is introduced namely Histogram Oriented Gradients Descriptor and Adabooster (HOGA) For Occlusion Handling and integrate this in a sliding window detection framework using HOG features and linear classification. The Proposed Tracking algorithm performs favorably against various methods that can be demonstrated by both qualitative and quantitative estimations challenging in a video sequences. The input for this project is live video. And the output is occlusion detection for the give video sequence.