Abstract: In this work it describes a complete framework for detecting objects in images and videos. Here the proposed approach is based upon ideas in computer vision, image processing and machine learning. These ideas provide an easy to use, simple and fast method for object detection. The main objective of this project is developing robust images feature sets for the task of object detection. And well normalized grids of gradient orientation histograms are used for proposed feature sets. These features helps to provide invariance in object location, changes in shape and good resistance to illumination changes and shadowing, background clutter and camera view point. For automatic/semi-automatic face detection here face recognition algorithms and motion detection algorithm are integrated. Here face detection technology is used to sort faces by their similarity to a chosen face or live webcam reducing user workload to search faces that belongs to the same person. Even many progress made in recent years, face recognition is a challenging topic in computer vision research.
Keywords: HOG, LBP, face recognition, feature sets.
Title: HOG Based Face Detection in Live Video Streaming
Author: Detty Susan George
International Journal of Computer Science and Information Technology Research
ISSN 2348-1196 (print), ISSN 2348-120X (online)
Research Publish Journals
