Visual Recognition by Object Centric And Scene-Centric Methods

Computer vision is a field that includes methods for acquiring, processing, analyzing and understanding of images and also has been described as the enterprise of automating and integrating a wide range of processes and representations for vision perception. Visual recognition is one of the major problems in computer vision. It includes the problems of scene classification, image annotation, image retrieval, object recognition and object detection. Context is a rich source of information about an object identity, location and scale. A novel frame work to context modeling is based on the probability of co-occurrence of objects and scene is proposed. Images are represented by their posterior probabilities with respect to set of contextual models build upon the Bag of-Features image representation. Representing images by posterior probabilities are remarkably noise-free and an effective model of the contextual relationships between semantic concepts. This modeling is based on two classes of representation. The first class consists of methods that model contextual relationships between sub-image entities. Methods in the second class adopt a scene- centric representation. Context models are learned from.Assistant Professor, Department of Information Technology.. PA College of Engineering & Technology, E-mail: suvethame@gmail.com Professor and Head, Department of Computer Science & Engineering, PA College of Engineering & Technology, E-mail: chitrapacet@gmail.com.entire images by generating a holistic description of the scene or its gist.