While we could post these on our publications page, we feel that they deserve a page of their own. Abstract: Semantic labeling is the task of assigning category labels to regions in an image.
For example, a scene may consist of regions corresponding to categories such as sky, water, and ground, or parts of a face such as eyes, nose, and mouth.
Thomas Heseltine Ph D Research Student Advanced Computer Architecture Group Department of Computer Science The University of York Tel: 44 (0)1904 432722 Email: [email protected] (ps) (Word) We explore research carried out in the field of face recognition, with the ultimate aim of producing a highly effective face recognition algorithm, for use in such application areas as secure site access, suspect identification and surveillance. 3D Face Recognition - Graph Matcher (login required).
A new line of research is proposed to analyse and compare the advantages offered by the various 2D (intensity image) approaches and newly emerging 3D (geometrical surface structure) approaches. Current research also includes the following (information regarding which, will become available shortly): Face Recognition: Two-dimensional and three-dimensional techniques Ph D Thesis (pdf, 4.39MB) Face Recognition: A Literature Review Research area literature review in Power Point (1.61MB) Evaluation of Image Pre-processing Techniques for Eigenface Based Face A research paper evaluating the improvements gained by use of various image pre-processing techniques, when applied to the Eigenface based method of face recognition.
The goal of this thesis is to develop methods for improving scene text recognition.
We do this by incorporating new types of information into models and by exploring how to compose simple components into highly effective systems.Typical approaches for this task include the conditional random field (CRF), which is well-suited to modeling local interactions among adjacent image regions.However the CRF is limited in dealing with complex, global (long-range) interactions between regions in an image, and between frames in a video.We develop and test both 2D and 3D face recognition systems on a large database of subjects and demonstrate how simple image pre-processing methods can significantly improve performance of existing 2D approaches. Current results gathered from tests of our 3D face recognition system indicate that the approach is capable of achieving significantly lower error rates than existing 2D systems. In the second part, we incorporate unsupervised feature learning based on convolutional restricted Boltzmann machines to learn a representation that is tuned to the statistics of the data set.We show how these features can be used to improve both the alignment quality and classification performance.In the third part, we present a nonparametric Bayesian joint alignment and clustering model which handles data sets arising from multiple modes.We apply this model to synthetic, curve and image data sets and show that by simultaneously aligning and clustering, it can perform significantly better than performing these operations sequentially.The RBM is a generative model which has demonstrated the ability to learn the shape of an object and the CRBM is a temporal extension which can learn the motion of an object.Although the CRF is a good baseline labeler, we show how the RBM and CRBM can be added to the architecture to model both the global object shape within an image and the temporal dependencies of the object from previous frames in a video.