Object Detection in Videos using Shot Clustering

This is a new approach to object detection and tracking are the challenging task in many computer vision applications. The clustering is a natural solution to abbreviate and organize the content of video. Research on object detection suggests that momententous objects can be well represented as more feature extraction from appropriately chosen video. Tracking is usually performed in the context of higher-level applications that require the location and/or shape of the object in every frame. In proposed system use shot clustering which is based on feature extraction. Object detection and tracking in motion based videos by using the joint color and texture histogram to represent a target. Apart from conventional color histogram feature of the object are also extracted by using LBP technique to represent the object. All frames are sequentially aligned for displaying tracking result compared with traditional color histogram method. Implemented algorithm extracts effectively features in the target region which represents more robustly than tracking produced in other similar method. For future investigation is to multitracking of detected object in videos and more number of feature extractions.