Abstract---Person re-identification is an approach of identifying the person at different locations and time across camera views in surveillance system. Person re-identification probably the open challenge for low-level video surveillance in the presence of a camera network. As person move from one camera view to another camera view the same person countered as two different persons .In order to reduce false count and enable seamless tracking, we proposed model- free gait representation. In this approach, width vector profile and width vector mean are taken as features. Applying of Normalization on features helps to achieve the results more accurately. To solve classification problem different distance metrics are used. The Experiments are carried out on CASIA gait database of gait dataset A. Re-identification results provided for the normalization of features and without normalization of features, results recorded for different view angles with respect to camera and results of applied different distance metrics are provided. Keywords—person re-identification, width vector profile, width vector mean, normalization. I. INTRODUCTION Person Identification or recognition has been receiving broad interests and it is highly desirable in applications such as security monitoring, authentication, etc. In order to recognize a person, different traits, including fingerprint, face, iris and gait can be used. Among these possible traits, face and body are preferred since they can be acquired without the person's cooperation. Using of human operator manual re-identification in large camera networks are expensive and inaccurate. Now a days many people using Gait recognition system since it has unique advantages as compared with other biometrics. Gait recognition is a task to identify or verify Indiduals by the way they walk shown in Fig.1.In video surveillance based application identifying the human gait is Important because it captures the human from a distance. So we choose gait based approach for person re-identification. Fig.1. Gait Cycle Gait recognition methods mainly classified into two major methods [1]; model-based and model-free methods. Model-based methods obtain series of static or dynamic body parameters via modeling or tracking body components such as limbs, legs, arms and thighs. View-invariant and scale- independent are main advantages of model-based approach. But model-based approaches are sensitive to the quality of gait sequences to achieve high accuracy and their computational cost also high due to it’s large parameter calculations. Model-free approaches focus on either shapes of silhouettes or the whole motion of human bodies. Model-free approaches are insensitive to the quality of silhouettes and also have the advantage of low computational costs. A.Overview of the proposed method The structure of proposed gait recognition system is in Fig.2.The system consist of mainly of three modules: One is preprocessing unit, here we detect the human movements and then extract the binary silhouette image from each frame. Background is eliminated from each image. The second feature extraction unit, extract the width vector features from normalized silhouette images. In the third step for the Person Re-Identification across Multiple Camera Views V Ramya #1 , V V Vineela #2 ,V Srinivasa Rao #3 ,K Srinivas #4 # Department of Computer Science and Engineering, VR Siddardha Engineering College, Vijayawada, Andhra Pradesh, India. 1 ramyarupav@gmail.com 2 venkatavineela3@gmail.com 3 drvsrao9@gmail.com 4 vrdrks@gmail.com * Software Group, Advanced Data Processing Research Institute– ISRO, Hyderabad, Telangana, India. Vol. 14 ICETCSE 2016 Special Issue International Journal of Computer Science and Information Security (IJCSIS) ISSN 1947-5500 [https://sites.google.com/site/ijcsis/] 172 Proceedings of 3rd International Conference on Emerging Technologies in Computer Science & Engineering (ICETCSE 2016) V. R. Siddhartha Engineering College, Vijayawada, India, October 17-18, 2016