Unleash Your Potential - Namagunga Girls Coding Club
Face Recognition by Sumudu Ranasinghe
1. Research Paper Analysis Ranasinghe A.A.S.P UWU/IIT/08/033 Independent Study and Seminar IIT362-1 Industrial Information Technology Uva Wellassa University Of Sri Lanka
4. Introduction Face recognition has become a popular area of research in computer vision and one of the most successful applications of image analysis and understanding. A set of two task: Face Identification: Given a face image that belongs to a person in a database, tell whose image it is. Face Verification: Given a face image that might not belong to the database, verify whether it is from the person it is claimed to be in the database.
6. Face Detection + Recognition Detection accuracy affects the recognition stage Key issues: Correct location of key facial features (e.g. the eye corners) False detection Missed detection
7. DIFFERENT APPROACHE Describe the different methods of face recognition. Feature extraction methods Holistic methods Hybrid methods
8. Feature extraction methods Feature extraction is the task where we locate facial features, Eg: the eyes, the nose, and the chins etc. This task may be performed after the face detection task Or recognition time. big challenge for feature extraction methods is feature “restoration“. Facial features are invisible according to the large variation.
9. Feature extraction methods This method is widely used to create individual vectors for each person in a system, the vectors are matched when an input image is being recognized.
11. Holistic methods Holistic methods uses the whole face region as the input to a recognition system. focuses a holistic method using eigenfacesto recognize still faces.
12. Face Recognition Using Eigenfaces The first stage is to insert a set of images into a database, these images are called the training set, this is because they will be used when we compare images and when we create the eigenfaces. The second stage is to create the eigenfaces. Eigenfacescan now be extracted from the image data by using a mathematical tool called Principal Component Analysis (PCA). When the eigenfaceshave been created, each image will be represented as a vector of weights. The system is now ready to accept incoming queries.
13. Face Recognition Using Eigenfaces The weight of the incoming unknown image is found and then compared to the weights of those already in the system. If the input image's weight is over a given threshold it is considered to be unknown. The identification of the input image is done by finding the image in the database whose weights are the closest to the weights of the input image. The image in the database with the closest weight will be returned as a hit to the user of the system.
14. Hybrid methods Hybrid face recognition systems uses a combination of both holistic and feature extraction methods. Hybrid method of face recognition by using 3D morphable model. The model makes it possible to change the pose and the illumination on the face.
15. 3D morphablemodel Took face recognition to a new level. By being able to use a morphable3D model to create synthetic images has proven to give good results. It is a very applicable approach that solves many of the problems. system achieved a recognition rate of 90%.
16. Problems of Face Recognition when comparing a database image with an input image. The main concern is of course that all images of the same face are heterogeneous. When image databases are created they contain good scenario images. concerning deferent facial expressions as well. The system must be able to know that two images of the same person with deferent facial expressions actually is the same person. makeup, posing positions, illumination conditions, and comparing images of the same person with and without glasses.